University of Southampton Research Repository …...University of Southampton Faculty of Natural and Environmental Sciences School of Chemistry Doctor of Philosophy High-throughput

University of Southampton Research Repository

ePrints Soton

Copyright © and Moral Rights for this thesis are retained by the author and/or other copyright owners. A copy can be downloaded for personal non-commercial research or study, without prior permission or charge. This thesis cannot be reproduced or quoted extensively from without first obtaining permission in writing from the copyright holder/s. The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the copyright holders.

When referring to this work, full bibliographic details including the author, title, awarding institution and date of the thesis must be given e.g.

AUTHOR (year of submission) "Full thesis title", University of Southampton, name of the University School or Department, PhD Thesis, pagination

http://eprints.soton.ac.uk

http://eprints.soton.ac.uk/

University of Southampton

Faculty of Natural and Environmental Sciences

School of Chemistry

High-throughput electrochemistry (HTP) - A new approach to

the rapid development of modified carbon electrodes

by

Aleksandra Pinczewska

A Thesis Submitted for the degree of

Doctor of Philosophy

December 2011

This thesis was submitted for examination in December 2011. It does not

necessarily represent the final form of the document deposited in the

university after examination

University of Southampton

Faculty of Natural and Environmental Sciences

School of Chemistry

Doctor of Philosophy

High-throughput (HTP) electrochemistry − A new approach to the rapid

development of modified carbon electrodes

by Aleksandra Pinczewska

Abstract

The major aim of this project was development of novel covalently modified glassy

carbon electrodes for application in NADH-dependent biosensors using

combinatorial and high-throughput methods. Studies on transition metal complexes

containing redox active 1,1-phenanthroline-5,6-dione (phendione) ligand(s) showed

they are effective electrocatalysts for oxidation of NADH. In order to covalently

tether the metal complexes at the GC surface, the design of GC electrodes modified

with novel metal complexes bearing phendione ligand(s) was proposed based on

sequential electrochemical and solid-phase synthesis methods. Initial work involved

optimisation of the process for modification of individual GC electrodes. Firstly,

following earlier work, the GC electrodes were electrochemically functionalised by

primary amines or a diazonium salt bearing Boc-protected amine groups, which

allowed introduction of chelating ligands at the GC surface under solid-phase

coupling conditions. The final step involved coordination of the bidentate ligand at

the GC surface to the metal centre and formation of novel metal complexes under

solid-phase coupling conditions. The successfully modified individual electrodes

were applied in the design of a library of GC electrodes modified with different

linkers, ligands and metal complexes and prepared in a combinatorial and parallel

way. The library was electrochemically screened in a high-throughput way using a

multichannel potentiostat, which allowed instant comparison of electrochemical and

electrocatalytic properties between different members of the library. The

experimental data extracted from HTP screening of the library were used for

evaluation of a) the surface coverage obtained for different library members; b) the

catalytic activity towards NADH oxidation and c) the kinetics parameters kcat and KM

for the electrocatalytic oxidation of NADH for all members of the library.

Table of Contents

1 Literature introduction .................................................................................................. 1

1.1 Principles of combinatorial and high-throughput methodology. .......................... 1

1.1.1 General ............................................................................................................... 1

1.1.2 Combinatorial synthesis methods. ..................................................................... 2

1.1.3 HTP screening strategies ................................................................................... 3

1.1.4 Application of combinatorial and HTP methods in electrochemistry. .............. 6

1.2 Chemically modified electrodes (CMEs) ............................................................. 9

1.2.1 Covalent modification of carbon electrodes .................................................... 10

1.2.2 Chemical functionalisation of the carbon electrode ........................................ 11

1.2.3 Electrochemical functionalisation of carbon electrodes .................................. 12

1.3 NADH co-enzyme and its electrochemical oxidation ........................................ 17

1.3.1 Electrochemistry of NADH/NAD+ .................................................................. 21

1.3.2 Mechanism of direct electrochemical NADH oxidation ................................. 22

1.3.3 NADH oxidation at modified electrodes ......................................................... 26

1.4 Metal complexes with 1,10-phenanthroline-5,6-dione as electrocatalysts for

NADH oxidation. ........................................................................................................... 33

1.5 Enzyme kinetics - the Michaelis Menten approximation ................................... 37

2 Functionalisation of individual GC electrodes ........................................................... 41

2.1 General ............................................................................................................... 41

2.2 Synthesis of ruthenium complexes with 1,10-phenanthroline-5,6-dione

chelating ligand. ............................................................................................................. 43

2.3 Attachment of bis-(dimethyl sulphoxide)(1,10-phenanthroline-5,6-dione)

ruthenium (II) chloride complex at the GC surface. ...................................................... 48

2.4 Attachment of (2,2’-bipyridine)(1,10-phenanthroline-5,6-dione) ruthenium

(II) chloride complex at the GC surface......................................................................... 62

2.5 Attachment of bis-(1,10-phenanthroline-5,6-dione)2 ruthenium (II) chloride

complex at GC surface. .................................................................................................. 69

2.6 Covalent modification of GC electrodes by (1,10-phenanthroline-5,6-

dione)zinc (II) chloride complex.................................................................................... 75

2.7 XPS analysis of the GC electrodes modified with ruthenium and zinc

complexes. ..................................................................................................................... 83

2.8 Modification of GC electrode by novel zinc complex containing pyridine

imidazole ligands. .......................................................................................................... 87

2.9 Conclusion .......................................................................................................... 91

3 Library of modified electrodes - preparation and HTP screening .............................. 95

3.1 Objectives ........................................................................................................... 95

3.2 Preparation and synthesis of the 63 modified electrode library. ........................ 99

3.3 HTP electrochemical screening of the library. ................................................. 107

3.4 Catalytic oxidation of NADH ........................................................................... 113

3.5 Control electrodes. ............................................................................................ 119

3.6 Kinetic analysis for 63 member library ............................................................ 121

3.6.1 General .......................................................................................................... 121

3.6.2 Theoretical kinetic model for NADH oxidation at modified electrodes.162 .. 123

3.6.3 Analysis of the heterogeneous rate constant kcat and the equilibrium

constant KM for the library of 63 modified electrodes. ............................................. 127

3.7 Conclusion ........................................................................................................ 137

4 Experimental Section ............................................................................................... 141

4.1 Synthesis ........................................................................................................... 141

4.1.1 General ........................................................................................................... 141

4.1.2 Instrumentation .............................................................................................. 141

4.1.3 Synthesis of 2-(2-(pyridin-2-yl)-1H-imidazol-1-yl)acetic acid ligand. ......... 143

4.1.4 Synthesis of ruthenium (II) complexes. ......................................................... 144

4.1.5 Synthesis of (2,2’-bipyridine)(1,10-phenanthroline-5,6-dione) ruthenium

(II) chloride.135 ........................................................................................................... 145

4.1.6 Synthesis of bis-(1,10-phenanthroline-5,6-dione)ruthenium (II)

chloride.135 ................................................................................................................. 146

4.2 Electrochemical and solid-phase covalent modification of GC electrodes. ..... 148

4.2.1 Instrumentation .............................................................................................. 148

4.2.2 Reagents ......................................................................................................... 148

4.2.3 Electrochemical modifications of GC electrodes. ......................................... 149

4.2.4 Attachment of (N-Boc-aminomethyl)benzene diazonium

tetrabluoroborate salt.176 ............................................................................................ 149

4.3 Solid-phase modification of the GC electrodes. ............................................... 150

4.3.1 General procedure for Boc removal of modified GC electrodes.176 .............. 150

4.3.2 General procedure for the coupling reaction of 2,2’-bipyridine-5-

carboxylic acid at the GC surface. ............................................................................ 150

4.3.3 General procedure for coupling of the 2-(1-(carboxymethyl)-1H-

imidazol-3-ium-2-yl)pyridin-1-ium trifluoroacetate salt at the GC electrodes. ....... 150

4.3.4 General procedure for grafting of ruthenium (II) complexes at the

modified GC electrodes. ........................................................................................... 151

4.3.5 General procedure for grafting (1,10-phenanthroline-5,6-dione)zinc (II)

chloride complex at the GC surface. ........................................................................ 151

4.4 General procedure for electrochemical characterisation of the individual

modified GC electrodes. .............................................................................................. 152

4.5 Electrochemical screening of the library of 63 modified electrodes. ............... 153

5 Conclusion ................................................................................................................ 154

6 References ................................................................................................................ 157

DECLARATION OF AUTHORSHIP

I, Aleksandra Pinczewska, declare that the thesis entitled

High-throughput electrochemistry (HTP)-A new approach to the rapid development of

modified carbon electrodes and the work presented in the thesis are both my

own, and have been generated by me as the result of my own original

research. I confirm that:

� this work was done wholly or mainly while in candidature for a research

degree at this University;

� where any part of this thesis has previously been submitted for a degree

or any other qualification at this University or any other institution, this

has been clearly stated;

� where I have consulted the published work of others, this is always clearly

attributed;

� where I have quoted from the work of others, the source is always given.

With the exception of such quotations, this thesis is entirely my own work;

� I have acknowledged all main sources of help;

� where the thesis is based on work done by myself jointly with others, I

have made clear exactly what was done by others and what I have

contributed myself;

� initial parts of this work involving electrochemical covalent attachment of

different linkers have been published as:

Ghanem, M. A.; Chrétien, J.-M.; Pinczewska, A.; Kilburn, J. D.; Bartlett, P. N. J.

Mater. Chem. 2008, 18, 4917.

Signed: ………………………………………………………………………..

Date:…………………………………………………………………………….

AcknowledgmentAcknowledgmentAcknowledgmentAcknowledgment

Firstly, I am truly indebted and thankful to my supervisors Prof. Phil Bartlett and Prof.

Jeremy Kilburn for giving me an opportunity to work on the Phd project where I

gained and developed broad range of synthetic and electrochemical skills. Phil, thanks

for all your patience and understanding when explaining all the electrochemical

problems and your enthusiasm for the three years. Jeremy, many thanks for your

patience and understanding when the things did not turn out right in the lab and your

advice on organic and synthetic problems and also for giving me an opportunity and

trust to carry on the HTP project in QMUL.

I would like to thank my advisor, Dr Martin Grossel, for his advices and overall

enthusiasm during my three years of Phd study. Prof. Gill Reid for her valuable advice

on synthesis of metal complexes and being so supportive whenever I have asked for

help and advice.

I would like to thank past and present members of the research project, Jean-Mathieu

Chretien and Mohammed Ghanem for introducing me into the project and their huge

patience and support during my learning process. Also Dr Sally Dixon, for her

priceless advice on synthesis and guidance during my synthetic work and for setting a

great example as a researcher. Thanks to Dr Maciej Sosna for his help, involvement

and enthusiasm in latter stage of the work with the HTP library.

I have been extremely lucky to work with both electrochemical and synthetic research

groups where I met fantastic people and this is why I thank following friends:

Biniam, Emma, Will and other present and past member of Jeremy’s group for being

such lovely and understanding friends and for their help with different problems in the

synthetic lab.

Thanks to fantastic and enthusiastic people in Phil’s group who helped me understand

many aspects of practical electrochemistry and were always happy to help with

everyday trouble shooting in the lab and answer my endless electrochemical

questions.

Finally, I would like to say how much I appreciate the support and help I have

received from my family, and most importantly Ross, who supported me during all

this time and always believed in me. I bet you would be very happy to read this

acknowledgment because it means I have finished writing!!

I have met many fantastic people in Southampton, who were not only my research

mates, but also became dear friends. Amongst them was also Angie, Magda and Ala

and I do really appreciate all your support and the precious advice I have received

from you girls.

List of Symbols

Symbol Description Unit

A

geometrical electrode area

cm2

D diffusion coefficient cm2 s-1

E°’ formal redox potential V

Ea potential of anodic peak V

Ebind binding energy eV

Ec potential of cathodic peak V

dEp separation peak potential V

Emp middle peak potential V

F Faraday constant C mol-1

ρ surface roughness factor -

Γmed surface coverage of mediator mol cm-2

i Current A

icat catalytic current A

inorm normalised catalytic current A

j current density A cm-2

n number of electrons -

kcat catalytic reaction rate s-1

kD mass transport rate constant cm s-1

KD dissociation constant mol cm-3

kE rate of electrochemical regeneration of mediator at the GC surface s-1

kME effective electrochemical rate constant cm s-1

KME effective Michaelis-Menten equilibrium constant mol cm-3

KM Michaelis-Menten equilibrium constant mol cm-3

pH −log10[H+] -

R gas constant J K mol-1

Q charge density C cm-2

T Temperature K

ν scan rate V s-1

V kinematic viscosity cm2 s-1

Vo kinematic initial viscosity cm2 s-1

Vmax kinematic maximum viscosity cm2 s-1

Common Abbreviations

Symbol Description

Ag/AgCl

silver/silver chloride reference electrode

BA benzylamine linker

BDA 1,4-butanediamine linker

CDCl3 deuterated chloroform

CME chemically modified electrode

CV cyclic voltammetry

DCM dichloromethylene

DIPEA diisoproylethylamine

DMF N, N-dimethylformamide

DMSO-d6 dimethylsulphoxide dueterated

EDA 1,2-ethylenediamine

EDDA 2,2'-(ethylenedioxy)diethylamine linker

ES+ electron spray positive ionisation

EtOH Ethanol

GC glassy carbon

HBTU O-(benzotriazol-1-yl)-N,N,N’,N’-tetramethylammonium

hexafluorophosphate

HCl hydrochloric acid

HDA 1,6-hexanediamine linker

HR-MS high resolution mass spectroscopy

HTP high-throughput

IR infrared spectroscopy

LR-MS low resolution mass spectroscopy

MeCN acetonitrile

MS-MALDI matrix assisted laser/desorption ionisation mass spectroscopy

NMR nuclear magnetic resonance

NAD+ nicotine adenine dinucleotide in oxidised form

NAD∞+ nicotine adenine dinucleotide in oxidised form in bulk solution

NADo+ nicotine adenine dinucleotide in oxidised form at electrode-solution

interface

NADH nicotine adenine dinucleotide in reduced form

NADH∞ nicotine adenine dinucleotide in reduced form in bulk solution

NADHo nicotine adenine dinucleotide in reduced form at electrode-solution

interface

[NADH-Q] NADH and mediator intermediate complex at the electrode surface

phendione 1,10-phenanthroline-5,6-dione

[Q] oxidised form of mediator at the electrode surface

[OH] reduced form of mediator at the electrode surface

SCE standard calomel reference electrode

TBATFB tertbutyl ammonium tetrafluoroborate salt

TFA trifluoroacetic acid

XDA p-xylene diamine linker

XPS X-ray photoelectron spectroscopy

Index of Figures

Figure 1.1 Split-pool synthesis to prepare combinatorial libraries of compound

mixtures; spheres represents resin beads; A, B and C represents the set of building

blocks; border are the reaction vessels. In the case where two building blocks are

used, in each coupling step after two stages, a total number of 9 different compounds

are formed, one on each resin bead. ............................................................................. 3

Figure 1.2 Process for the iterative deconvulation screnning of the combinatorial

library containing variable building blocks A, B, C and D; compounds placed in

brackets are hits obtained after each iterative step. ...................................................... 4

Figure 1.3 Positional scanning deconvulation, where 16 sublibraries (pools) with 64

members each were screened at the same time. ........................................................... 5

Figure 1.4 Functionalisation of the carbon surface by chemical oxidation followed by

conversion of the carboxylic acids into more active acyl chlorides........................... 11

Figure 1.5 Chemical functionalisation of the carbon surface by hydroxyl groups

followed by attachment of organosilanes. .................................................................. 12

Figure 1.6 Electrochemical functionalisation of electrodes surface by applying Kolbe

reaction at metal and carbon electrodes. .................................................................... 13

Figure 1.7 Electrochemical functionalisation of carbon electrode by oxidation of

primary amines.64 ....................................................................................................... 13

Figure 1.8 The general mechanism of an enzymatic reaction catalysed by

NAD(P)+/NAD(P)H co-enzyme. ............................................................................... 18

Figure 1.9 Structures of NAD and NADP and their reversible redox reaction by

hydride ion transfer between oxidised form of NAD(P)+ and reduced form of

NAD(P)H carried out at C-4 of the enzymatically active nicotinamide ring;

substituent R represents enzymatically inactive linkage consists of ribose, adenine

base and phosphate groups and substituent X at one of the ribose rings is H for NAD

and a phosphate group PO32- for NADP. .................................................................... 18

Figure 1.10 Stereospecific hydride transfer during redox reaction between NAD(P)+

and NAD(P)H. ............................................................................................................ 19

Figure 1.11 Schematic diagrams illustrating the Rossmann folding of β-strands and

α-helices in the NAD-binding protein domain of dehydrogenase enzyme.83 ............ 20

Figure 1.12 Proposed reaction pathway for the electrochemical oxidation of NADH

according to the ECE mechanism; kH+ is a deprotonation rate constant. ................... 24

Figure 1.13 General scheme of the possible electron transfer pathway during NADH

oxidation mediated by surface oxygen redox systems; a) mediated by hydroxy

groups; b) mediated by molecular oxygen.73,89,97 ....................................................... 25

Figure 1.14 Possible proton transfer mechanism during oxidation of NADH

involving movement of a proton from C-4 of the nicotinamide ring to a proton

acceptor such as water molecule.73,89,97 ...................................................................... 25

Figure 1.15 The catalytic cycle for the oxidation of NADH by the mediator M

immobilised at the electrode surface.43,98 ................................................................... 27

Figure 1.16 Enzymatic reaction of the substrate S catalysed by NAD+ followed by

electrochemical reoxidation of the NADH by the mediator M at the rate constant

kobs, which is in kinetic competition with the rate constant kb of the enzyme

catalysed back reaction of NADH with the product P. .............................................. 28

Figure 1.17 Possible deactivation mechanism for the adsorbed o-quinone mediator

and NADH+· radical during electrochemical oxidation of the NADH.73,99 ................ 28

Figure 1.18 Proposed mechanism of the NADH oxidation catalysed by redox

mediator, X is a group working as a hydride acceptor; Y is an electron deficient

group.101,102 ................................................................................................................. 29

Figure 1.19 Examples of common redox dye mediators applied as effective catalysts

for NADH oxidation. ................................................................................................. 30

Figure 1.20 Structures of the organic conducting salts N-methylphenazinium

tetracyanoquinodimethane (NMP-TCNQ) as effective mediators for NADH

oxidation. .................................................................................................................... 32

Figure 1.21 a) Examples of the mediators consisting of the nitro functional groups; b)

electrochemical formation of the nitrosyl catalytic active form.115 ........................... 32

Figure 1.22 a) The reaction pathway of the phendione redox process in aprotic

solvent; a) Cyclic voltammogram for 1.2 mM solution of the phen-dione in MeCN

containing 0.1 M TBAP at graphite electrode.120 ...................................................... 33

Figure 1.23 a) General scheme of the reversible electrochemical redox reaction of the

phendione ligand in aqueous solvent; b) Cyclic voltammograms for 1.1 mM solution

of phendione in aqueous solution at pH 2.85 (curve B) and pH 6.8 (curve C) at a Pt

electrode and sweep rate 0.2 V s−1.120 ........................................................................ 34

Figure 1.24 Formation of the hydrogen bridge between 5,6-dihydroxy and imine

groups of the phendione species as proposed to explain of the irreversible cyclic

voltammograms observed at pH>5.119,122 ................................................................... 34

Figure 1.25 a) Structure of the metal complex of general form of

[M(phendione)3]2+; b) Typical cyclic voltammogram at a Pt electrode at scan rate of

20 mV s−1 for 0.5 M [Fe(phendione)3]2+ in aprotic solvent with 0.1 M TBAP,

reported by Abruña et al.120 ........................................................................................ 35

Figure 1.26 Relation between reaction velocity V and the concentration of substrate

defined by Equation 1.8. ............................................................................................ 40

Figure 2.1 Cyclic voltammogram of 1 mM solution of complex 3 in 0.1 M phosphate

buffer solution pH 7 at scan rate of 50 mV s−1 and GC electrode geometrical surface

area of 0.071 cm2. ....................................................................................................... 45

Figure 2.2 Cyclic voltammograms of 1 mM solution of complex 4 at GC electrode

(geometrical electrode area of 0.071 cm2); a) in 0.1 M phosphate buffer solution pH

7 at scan rate of 50 mV s−1; b) in acetonitrile with 0.1 M TBATFB at scan rate of 50

mV s−1. ....................................................................................................................... 47

Figure 2.3 Oxidation of linker 6 at the polished GC electrode with geometrical area

of 0.071 cm2; general reaction scheme and cyclic voltammogram (below) recorded in

10 mL of 15 mM solution of 6 in acetonitrile with 0.1 M TBATFB, from 0 to 2.25 V

vs. Ag/AgCl at scan rate of 50 mV s−1. ...................................................................... 49

Figure 2.4 Cyclic voltammograms obtained after heating of electrodes 7 (red) and 10

(black) in a solution of the complex 3 in 0.1 M phosphate buffer pH 7 at a scan rate

of 50 mV s-1 for electrodes with geometrical surface area of 0.071 cm2. .................. 51

Figure 2.5 Cyclic voltammograms of modified electrode 13 in the presence (red) and

absence (black) of 1 mM NADH in 0.1 M phosphate buffer solution pH 7 at a scan

rate of 50 mV s-1 (geometrical electrode area of 0.071 cm2). .................................... 54

Figure 2.6 Cyclic voltammogram recorded in 15 mM solution of linker 20 in

acetonitrile with 0.1 M TBATFB at scan rate of 50 mV s-1 at GC surface with

geometrical area of 0.071 cm2. ................................................................................... 56

Figure 2.7 Cyclic voltammograms of the modified electrodes 13 and 35-39 in the

presence (red) and absence (black) of 1 mM NADH in 0.1 M phosphate buffer

solution pH 7 at scan rate of 50 mV s-1 (geometrical electrode area of 0.071 cm2). .. 59

Figure 2.8 Values of Γmed calculated for modified electrodes 13 and 35-39; Each bar

represents values of Γmed calculated for single modified electrode according to

Equation 2.1. .............................................................................................................. 60

Figure 2.9 a) Catalytic current icat for modified electrodes 13 and 35-39 recorded in

1 mM NADH solution in 0.1 M phosphate buffer pH 7 at scan rate of 50 mV s-1; b)

Normalised catalytic currents inorm for the modified electrodes 13 and 35-39

calculated according to Equation 2.2; Each bar represents values of icat (a) or inorm

(b) calculated for a single modified electrode. ........................................................... 61

Figure 2.10 Cyclic voltammograms of electrode 41 recorded in the presence (red)

and absence (black) of 1 mM NADH solution in 0.1 M phosphate buffer pH 7 at a

scan rate of 50 mV s-1 and geometrical electrode area of 0.071 cm2. ........................ 63

Figure 2.11 Cyclic voltammograms recorded for the electrodes 40-45 modified with

complex B attached at the GC surface through six different linkers in presence (red)



Figure 2.12 Values of surface coverage Γmed of complex B calculated for modified

electrodes 41-46 based on experimental data obtained from the cyclic

voltammograms depicted in Figure 2.11. Each bar represents the value of Γmed

calculated according to Equation 2.1 for a single modified electrode. ...................... 66

Figure 2.13 a) Catalytic currents icat recorded for modified electrodes 41-46 in the

presence of 1 mM NADH in 0.1 M phosphate buffer solution pH 7 at a scan rate of

50 mV s-1; b) Normalised catalytic currents inorm calculated for electrodes 41-46

according to Equation 2.2. Each bar represents values of icat (a) or inorm (b) obtained

for single modified electrode. .................................................................................... 67

Figure 2.14 Attachment of complex 5 at electrode 10; a) 0.01 M of complex 5 in dry

DMF, 100 °C, 16 h; Cyclic voltammograms of electrode 47 recorded in the presence

(red) and absence (black) of 1 mM NADH solution in 0.1 M phosphate buffer pH 7

at a scan rate of 50 mV s-1 and geometrical electrode area of 0.071 cm2. ................. 70

Figure 2.15 Cyclic voltammograms recorded for electrodes 47-52 modified with

complex C attached at GC surface through six different linkers in the presence (red)



Figure 2.16 Values of surface coverage Γmed of complex C calculated for modified

electrodes 47-52 based on experimental data obtained from cyclic voltammograms

presented in Figure 2.15. Each bar represents value of Γmed calculated according to

Equation 2.1 for a single modified electrode. ............................................................ 73





for single modified electrode. ..................................................................................... 74

Figure 2.18 Preparation of electrode 55 under solid-phase reaction conditions based

on the literature procedure; a) silver nitrate, acetonitrile, complex 53, 16 h, 50 °C b)

0.01M complex 54 in acetonitrile, 50 °C, 16 h; Cyclic voltammogram of electrode 55

in phosphate buffer solution pH 7 at a scan rate of 50 mV s-1 and geometrical

electrode area of 0.071 cm2. ....................................................................................... 76

Figure 2.19 Optimisation process of formation of complex D at electrode 10 in 0.01

M solution of 53 in DMF a various reaction time and reaction temperatures; a)

Values of Γmed for complex D created at reaction times varying from 30 minutes to

16 h; b) Values of Γmed for complex D created at different reaction temperatures for

16 h. ............................................................................................................................ 78

Figure 2.20 Cyclic voltammograms recorded for electrodes 56-61 modified with

complex D attached at the GC surface through six different linkers in presence (red)



Figure 2.21 Values of surface coverage Γmed of complex D calculated for modified

electrodes 56-61 based on experimental data obtained from the cyclic

voltammograms in Figure 2.15. Each bar represents value of Γmed calculated

according to Equation 2.2 for a single modified electrode. .................................... 81





for single modified electrode. .................................................................................... 82

Figure 2.23 Examples of XPS photoelectron spectra recorded for GC sheets; a)

polished, blank GC sheet; b) with ruthenium complex A attached at the surface

through EDA linker; c) zinc complex D attached at the surface through EDA linker.

.................................................................................................................................... 85

Figure 2.24 Cyclic voltammograms of electrode 66 (black) and control electrode

(red) in 0.1 M solution of phosphate buffer pH 7 at a scan rate 50 mV s-1 and

geometrical electrode area of 0.071 cm2. ................................................................... 89

Figure 2.25 Cyclic voltammograms of electrode 66 in the presence (red) and absence

(black) of 1 mM NADH solution in 0.1 M phosphate buffer pH 7 at scan rate of 50

mV s-1 and geometrical electrode area of 0.071 cm2. ................................................ 90

Figure 3.1 Typical cyclic voltammograms of background currents recorded for blank

glassy carbon electrodes 13a-c with geometrical area of 0.71 cm2 in 0.1M phosphate

buffer solution pH 7 at scan rate of 50 mV s-1. ........................................................ 100

Figure 3.2 Typical cyclic volatammograms for linkers immobilisation at GC polished

electrodes in 15 mM solution of linker in MeCN with 0.1 M TABTFB at a scan rate

of 50 mV s-1; a) attachment of the EDA linker for electrode 13c; b) attchment of the

BA linker for electrode 39a. ..................................................................................... 101

Figure 3.3 Example of the parallel set up of reaction vessels used during solid-phase

synthesis of metal complexes at GC electrodes; the inset shows the reaction vessel

contain of solution of Zn(phendione)Cl2 and electrodes 56c, 58c, 59c and 61c. ..... 104

Figure 3.4 a) Effect of scan rate on modified GC electrode 13a in 0.1 M phosphate

buffer solution pH 7, electrode area of 0.71 cm2; b) Plot of anodic (red) and cathodic

(black) currents as a function of scan rate ν for modified electrode 13a. ................ 107

Figure 3.5 Average values of the separation pontential Ep (plot a) and the midpoint

potential Emp (plot b) observed for twenty different modifications in the library based

on the experimental data recorded in 0.1 M phosphate buffer solution pH 7 at scan

rate 50 mV s−1. ......................................................................................................... 110

Figure 3.6 Average values of the surface coverage Γmed of the metal complex

calculated for 20 different modifications in the library using Equation 2.1 based on

experimental data recorded in 0.1 M phosphate buffer solution pH 7 at scan rate of

50 mV s−1; geometrical electrode area of 0.071 cm2; Complexes A, B, D and E

contain one phendione ligand whereas complex C has two phendione ligands; a)

Each bar represents mean values of Γmed with the and standard error of the mean for

three replicate electrodes; b) three dimensional representation of the means. ......... 112

Figure 3.7 An example of a cyclic voltammograms recorded in the presence (black)

and absence (red) of 1 mM NADH in 0.1 M phosphate buffer solution pH 7 at a scan

rate of 50 mV s−1, geometrical electrode area of 0.071 cm2. Data shown in Figure 3.7

corresponds to a modified electrode 13a. ................................................................. 113

Figure 3.8 a) Means of the catalytic currents icat for 20 different modifications of the

library measured in 1 mM NADH solution at a scan rate of 50 mV s−1. Each bar

represents an average value of icat and the standard errors of the means calculated for

3 replicate electrodes; b) The means of icat values for 20 different modifications of

the library presented in three dimensional bar plot. ................................................. 114

Figure 3.9 Average values of the normalised catalytic currents inorm calcultated for 20

different modifications in the library based on the surface coverage of the phendione

ligands using experimental data recorded for 1 mM NADH in phopshate buffer

solution at scan rate of 50 mV s−1; a) Each bar represents an average value of inorm

and the standard errors of the mean calculated for 3 replicate electrodes; b) The

means of inorm values presented in a three dimensional bar plot. ............................. 116

Figure 3.10 Average values of the normalised catalytic currents inorm calcultated for

20 different modifications in the library based on the surface coverage of the metal

complexes using experimental data recorded for 1 mM NADH in 0.1 M phopshate

buffer solution at scan rate of 50 mV s−1. Complexes A, B, D and E contain one

phendione ligand whereas complex C has two phendione ligands; a) Each bar

represents an average value of inorm and the standard errors of the mean calculated for

3 replicate electrodes; b) The means of inorm values presented in a three dimensional

bar plot. .................................................................................................................... 117

Figure 3.11 Cyclic voltammograms for the modified electrode 13a and the control

electrodes 70a-c obtained duirng HTP screening in 0.1 M phosphate buffer solution

pH 7, scan rate of 50 mV s−1 (black curve), in 1 mM NADH aqueous solution at scan

rate of 50 mV s−1 (red curve).................................................................................... 119

Figure 3.12 Mechanism of NADH catalytic oxidation according to the Michaelis-

Menten model. .......................................................................................................... 121

Figure 3.13 Proposed reaction scheme for the oxidation of NADH at the mediator

modified electrodes, according to Bartlett and Wallace.163 Subscripts ∞ and O

indicate NAD in the bulk solution and at the surface/solution interface. Q represents

the oxidised form of the mediator; QH represents the reduced form of the mediator;

[NADH-Q] represents the intermediate complex; kD is the diffusion rate constant and

is considered to have the same values for NADH and NAD+; KM is the Michaelis-

Menten equlibrium constant for the complex [NADH-Q]; kcat is heterogenous rate

constant for the chemical reaction within [NADH-Q] complex; kE is the rate constant

for the electrochemical oxidation of the reduced mediator QH. .............................. 123

Figure 3.14 Plot of catalytic current icat as a function of NADH concentration at a

scan rate 10 mV s−1 obtained for modifed electrodes 13a-c; the points represent the

experimental data and the lines are the best fits calculated from the theoretical model

using Equation3.7. .................................................................................................... 127

Figure 3.15 a) Plot of catalytic current as a function of NADH concentration at scan

rates of 10, 20, 50 and 100 mV s−1; the points are experimental data, the lines are

fitting curves obtained calculated from Equation 3.7; b) Comparison of the fitting

curves obtained for scan rate 100 mV s−1 with experimental data in 2 mM NADH

(black line) and without 2 mM NADH (red line). .................................................... 128

Figure 3.16 Average values of catalytic rate constants kcat caclulated using surface

coverage of the phendione ligands obtained for 20 different modifications in the

library, where kcat for each individual electrode was calculated using experimental

data of 12 single sets of HTP measurments performed in 0.1 M phosphate buffer

solution pH 7; complexes A, B, D and E contain one phendione ligand whereas

complex C has two phendione ligands; a) means values of kcat with the standard error

of the mean of three replicate electro`des; b) 3D bar plot of average values of kcat for

20 different modifications in the library. ................................................................. 130

Figure 3.17 Average values of catalytic rate constants kcat caclulated using surface

coverage of the metal complex obtained for 20 different modifications in the library.

The value of kcat for each individual electrode was calculated using experimental data

for 12 single sets of HTP measurments performed in 0.1 M phosphate buffer solution

pH 7; a) mean values of kcat with the standard error of the mean for the three replicate

electrodes; b) 3D bar plot of average values of kcat for 20 different modifications in

the library. ................................................................................................................ 131

Figure 3.18 Average values of Michaelis-Menten constnat KM and their three

dimensional representation obtained for 20 different modifications in the library.

Each bar represents mean value of KM and standard error of the mean of three

replicate electrodes, where KM for each individual electrode was calculated using

experimental data of 12 single sets of HTP measurments performed in 0.1 M

phosphate buffer solution pH 7. ............................................................................... 132

Figure 3.19 Average values of rate constants kcat/KM at low NADH concentrations

([NADH] < KM) obtained for 20 different modifications in the library; Each bar

represents mean values of kcat/KM with a standard error of the mean for three replicate

electrodes; a) Means for kcat/KM calculated using values of kcat presented in Figure

3.16 for the surface coverage of phendione ligands and KM in Figure 3.18; b) Means

for kcat/KM calculated using values of kcat presented in Figure 3.17 for the surface

coverage of metal complexes and KM in Figure 3.18. .............................................. 134

Index of Schemes

Scheme 1.1 a) Conventional synthesis: one building block A reacts with one reagent

B; b) combinatorial synthesis: each of the building blocks of series A1-n reacts with

each of the building blocks of series B1-n at the same time to give series of products

AnBn. ............................................................................................................................ 2

Scheme 1.2 General representation of an electrochemical redox reaction catalysed by

a mediator. .................................................................................................................... 9

Scheme 1.3 General scheme of the GC electrodes modified by various diamine

linkers and the anthraquinone redox centre.57 ............................................................ 14

Scheme 1.4 Electrochemical covalent attachment of primary alcohols at the carbon

surface. ....................................................................................................................... 14

Scheme 1.5 Electrochemical functionalisation of carbon electrodes by reduction of a

diazonium salt. ........................................................................................................... 15

Scheme 1.6 Covalent functioanlisation of the GC electrodes by 4-(N-Boc-

aminomethyl)benzene diazonium tetrafluoroborate salt linker and AQ redox centre;

a) from 0.6 to −1 V vs. Ag/AgCl in 0.1 M TBATFB in MeCN; b) 4 M HCl in

dioxane; c) Anthraquinone-2-carboxylic acid (AQ), HBTU, DIEA, DMF, R.T., 16

h.55 .............................................................................................................................. 16

Scheme 2.1 General approach for sequential covalent modification of the GC

electrodes developed by Kilburn and Bartlett;57 a) electrochemical reduction of Boc

protected amine diazonium salts or oxidation of mono-Boc protected primary

diamine; b) removal of the Boc protecting group; c) solid-phase coupling reaction of

a redox center at the GC surface. ............................................................................... 41

Scheme 2.2 Proposed strategy of covalent immobilisation of metal complexes at the

GC electrode using electrochemical and solid-phase coupling conditions. ............... 42

Scheme 2.3 Preparation of the precursor 1 and its X-ray crystallographic structure

(right); a) 5 min reflux in neat DMSO under nitrogen, acetone, 52 %. ..................... 43

Scheme 2.4 Preparation of complex 3; a) precursor 1, ethanol, reflux, 14 h, 60 %. .. 44

Scheme 2.5 Synthesis of the complex 4; a) 2,2’-bipyridine, DMF, reflux 4 h, reduced

light, 60 %. ................................................................................................................. 46

Scheme 2.6 Preparation of complex 5; a) precursor 1, DMF, reflux 4 h, 33 %. ........ 47

Scheme 2.7 Removal of Boc protecting group from electrode 7 followed by solid-

phase coupling of the ligand 9; a) 4.0 M hydrochloric acid in 1,4-dioxane, 1 h, b)

HBTU, DIPEA, DMF, 16 h, room temperature. ........................................................ 50

Scheme 2.8 Optimisation process of the attachment of the complex 3 at the modified

electrode 10 under solid-phase reaction conditions. .................................................. 53

Scheme 2.9 Covalent attachment of linkers 6 and 18-21 at the GC electrodes; a)

applied potential from 0 to 2.25 V vs. Ag/AgCl in 15 mM solution of the linker in

acetonitrile with 0.1 M TBATFB at scan rate of 50 mV s-1. ...................................... 55

Scheme 2.10 Attachment of linker 27 at the GC electrode; general scheme and cyclic

voltammogram of the electrochemical reduction of 15 mM solution of 27 in

acetonitrile with 0.1 M TBATFB at scan rate of 50 mV s-1 at GC electrode with

geometrical electrode area of 0.071 cm2. ................................................................... 57

Scheme 2.11 Solid-phase modifications of the electrodes 22-26 and 28; a) 4.0 M

hydrochloric acid in 1,4-dioxane, b) ligand 9, DMF, HBTU, DIEA, room

temperature, 16 h, c) complex 3, DMF, 100 °C, 16 h. ............................................... 58

Scheme 2.12 Optimisation process of the synthesis of complex B at modified

electrode 10. ............................................................................................................... 63

Scheme 2.13 Attachment of complex 4 at electrodes 29-34 functionalised by six

different linkers; Each modified electrode 41-46 was prepared individually under the

same solid-phase reaction conditions a) 0.01 M of complex 4 in dry DMF, 100 °C,

16 h. ............................................................................................................................ 64

Scheme 2.14 Attachment of complex 5 at electrodes 29-34; Each modified electrode

47-52 was prepared individually under the same solid-phase conditions a) 0.01 M of

complex 5 in dry DMF, 100 °C, 16 h. ........................................................................ 71

Scheme 2.15 Attachment of complex 53 at electrodes 29-34, functionalised by six

different linkers; Each modified electrode 56-61 was prepared individually under the

same solid-phase reaction conditions a) 0.01 M of complex 51 in dry DMF, 50 °C,

16 h. ............................................................................................................................ 79

Scheme 2.16 Synthesis of ligand 64; a) 2.5 M n-buthylithium in hexane, 2-bromo-

tertbutyl acetate THF, 78 °C, overnight, 76 %; b) 20 % trifluoroacetic acid (TFA) in

DCM, overnight, R. T., 98 %. .................................................................................... 88

Scheme 2.17 Preparation of modified electrode 66; a) 1 M solution of 60 in DMF, 10

equiv. of DIPEA, HBTU, R.T., 16 h; b) 0.01 M solution of 53 in DMF, 50 °C, 16 h.

.................................................................................................................................... 88

Scheme 3.1 General approach to covalent modification of GC electrodes by metal

complexes. .................................................................................................................. 95

Scheme 3.2 Attachment of linkers 9 and 64 at the functionalised GC surface; a)

ligand 9, HBTU, DIPEA, DMF, R.T., 16 h; b) ligand 64, HBTU, 10 equiv. DIEA,

DMF, R.T., 16 h. ...................................................................................................... 102

Scheme 3.3 Solid-phase synthesis of metal complexes at the functionalised GC

electrodes; a) Ru(phendione)Cl2(DMSO)2, DMF, 100 °C, 16 h. b)

Ru(phendione)(bipy)Cl2, DMF, 100 °C, 16 h, c) Ru(phendione)2Cl2, DMF, 100 °C,

16 h, d) and e) Zn(phendione)Cl2, DMF, 50 °C. ...................................................... 103

Index of Tables

Table 2.1 The values of approximate atomic ratios Ru/N for the GC sheets modified

with six different linkers and four different metal complexes calculated from XPS

experimental data (italics). Values of Ru/N* represents theoretical atomic ratios

Ru/N based on assumption that each linker at the surface is coupled with a metal

complex. ..................................................................................................................... 86

Table 3.1 Design of the library of 63 modified electrodes; the columns represent the

four linkers: EDA (1,2-ethylenediamine), HDA (1,6-hexanediamine), EDDA (2,2'-

(ethylenedioxy)diethylamine) and BA (p-benzylamine); the rows represent: the

ruthenium complexes (A, B and C), the zinc complexes (D, E) and the control

electrodes 70a-c. ......................................................................................................... 97

Table 3.2 Arrangment of electrodes and reaction vessels during parralel attachment

of the metal complexes............................................................................................. 105

Table 3.3 Summary of anodic and cathodic potentials peaks Ea and Ec, separation

peak ∆∆∆∆Ep, midpoint potential Emp = (Ea + Ec) / 2 obtained from cyclic

voltammograms recorded in 0.1 M phosphate buffer solution pH 7 at a scan rate of

50 mV s−1. ................................................................................................................ 108

1 Literature introduction

1.1 Principles of combinatorial and high-throughput methodology.

1.1.1 General

Combinatorial and high-throughput (HTP) techniques are powerful tools widely used

in development and evaluation of biological systems, pharmaceuticals, catalysts and

many classes of inorganic compounds.1-4 The combinatorial and HTP method

involve design, synthesis and rapid evaluation of large numbers of structurally and

compositionally diverse compounds (libraries or arrays) in order to identify the

compounds or family of lead compounds, with the desired properties. Combinatorial

methodology was developed by Hanak,5 where the “multiple sample concept” was

applied for discovery of new materials. Since then, this method has been widely used

in the pharmaceutical industry, where the long and expensive drug discovery process

was accelerated by rapid and cost effective combinatorial and HTP strategies.

In general, the principles of combinatorial and HTP chemistry are as follows:

• Rapid synthesis of a large number of diverse new materials from different

combination of specific building blocks or processing conditions to create a

library of analogous products at the same time under identical reactions

conditions.

• Rapid evaluation of one or more chemical or physical properties of each

library member and selection of the lead compounds to be used for further

investigation.

• Further optimisation of leads varying the stoichiometries or structures

followed by their rapid evaluation, resulting in a more focused library of

compounds.

• Preparation of compounds with the best properties in quantities sufficient for

their detailed characterisation.

Literature introduction

2

1.1.2 Combinatorial synthesis methods.

The combinatorial synthetic approach is based on the preparation of many

compounds at the same time in a systematic manner, which is more effective than the

longer, classical, one compound at time method. As a result, the set of products of all

possible combinations of building blocks creates a combinatorial library (Scheme

1.1).

Scheme 1.1 a) Conventional synthesis: one building block A reacts with one reagent B; b)

combinatorial synthesis: each of the building blocks of series A1-n reacts with each of the building

blocks of series B1-n at the same time to give series of products AnBn.

In combinatorial synthesis, the total number of library members is given by N=b×x,

where b is the number of building blocks used in each reaction and x is the number of

reaction steps where new building blocks were added.

Combinatorial synthesis can be carried out in solution or on solid supports such as

resin beads, chips and pins. However, solid-phase conditions are more common due

to easier parallel work-up and high yield of the solid-phase reactions as one can use

excess of reagents in solution. The solid-phase condition has been widely used in

development of new organic compounds, particularly in peptide synthesis.3,6-8

Depending on the synthetic strategy and the library size, the library can contain of a

mixture of different compounds or separate, single substances. Two main

combinatorial synthesis methods exist:

• Parallel synthesis can be performed in solution or on solid supports, using

ordered arrays of separated reaction vessels according to the conventional

“one compound-one vessel” rule. The advantages of this method are the

control over the purity of the products. In addition, defined location of the


3

product in the array provides structure of the products prepared and that the

structure of the product is known from its defined location in the array. This

method can be applied to the synthesis of medium size of libraries (several

thousand compounds).

• The split-pool synthesis procedure allows for preparation of libraries of many

products using a few reaction vessels. In the first step, the resins are split

between, for example, three vessels and each resin is coupled with three

different building blocks A1, A2 and A3. After this step, all the three products

are pooled together in one vessel where common steps such as deprotection

or resin washing are carried out. Analogous to the first step, the mix of resins

is divided into three vessels and each of them reacts with building blocks B1,

B2 and B3. The split and pool processes were repeated until the desired

combinatorial library is obtained. Each bead in the library is attached to just

one single compound.

Figure 1.1 Split-pool synthesis to prepare combinatorial libraries of compound mixtures; spheres

represent resin beads; A, B and C represents the set of building blocks; border are the reaction vessels.

In the case where two building blocks are used, in each coupling step after two stages, a total number

of 9 different compounds are formed, one on each resin bead.

1.1.3 HTP screening strategies

Characterisation of combinatorial libraries can be carried out one by one using

conventional analytical methods or by using high-throughput parallel screening.

Analytical methods such as mass spectroscopy, NMR and IR are applicable for

screening of single substances from the library. In the case of larger libraries of


4

mixtures of compounds, accurate structural determination by analytical methods is

difficult due to the similarity of the structures and properties of the evaluated

compounds. Therefore, in order to determine the lead compound(s) in the

combinatorial library different screening strategies were developed.9

For combinatorial libraries which comprise of a mixture of up to several

thousand compounds, deconvulation screening methods can be applied for

determination of the structure and biological activity of the hit compound(s).9-13

Probably the most straightforward deconvolution strategy is iterative deconvolution,

which has been used successfully in identifying several biologically active

compounds.14 This strategy is based on the synthesis of sub-libraries with compound

mixtures (pools) were one of the building blocks are fixed and the rest of the building

blocks are set up in a combinatorial way (Figure 1.2). After screening, the pool with

the most desired properties is resynthesised with a second building block fixed. This

process is repeated until the combination of building blocks with the best properties

is obtained.

Figure 1.2 Process for the iterative deconvulation screnning of the combinatorial library containing

variable building blocks A, B, C and D; compounds placed in brackets are hits obtained after each

iterative step.

Although iterative deconvulation approach has found a few practical applications, the

rate of hit compound discovery is limited by the requirement for several rounds of

the synthesis and testing.10,11,13 To overcome this, HTP deconvulation screening may

be performed by positional screening, where all the possible combinations of the


5

building blocks are prepared at once (Figure 1.3).9 Each of the sub-libraries contains

one of defined building blocks at one position and a mixture of building blocks at the

others. The number of sub-libraries is the same as the number of variable positions in

the substitution pattern. This method of HTP screening is faster and allows screening

of all of the possible combinations of building blocks in one measurement. However,

this also requires preparation of larger libraries. Screening of the library all at once

allows for direct deduction of hit compounds.

Figure 1.3 HTP positional scanning deconvulation method, where each of 16 sublibraries (pools) is

containing one defined building block A1-4 -D1-4 were screened at the same time.

In the case of combinatorial libraries comprising mixtures of solid-bound

compounds, the compounds can be screened still attached to the solid support using

“on-bead” screening.9

This method works particularly well in the case of compounds with biological

activity, where the solid-bound library is treated with a labelled soluble biological

target. Example of the “on-bead” screening strategy includes employing fluorescent

label receptors, which bind to the compounds with the highest affinity for biological

receptors.15,16 The “on-bead” screening is particularly useful for libraries of several

thousand to a million compounds and isolation of a few bioactive compounds from

many inactive ones.

The HTP screening of combinatorial libraries using the general strategies such as

deconvulation and “on-bead” attracted significant interest mostly in development of

pharmaceuticals and biologically active compounds.17 The combinatorial and HTP


6

methodologies have also been applied for development of new solid-state materials,

including superconducting materials, magnetoresistants, zeolites or polymers.2,18-21

Significant advances in the development of combinatorial and HTP approaches have

been also observed in discovery of new catalysts,22-29 including asymmetric catalysts

for organic synthesis,2,26 metal binding peptidic ligands22,29 or metal complexes27 as

enzyme mimetics. Common HTP techniques for screening the combinatorial libraries

towards development of new catalysts and materials include mass spectrometry,

circular dichroism, IR thermography as well as increasing number of electrochemical

techniques.2

1.1.4 Application of combinatorial and HTP methods in electrochemistry.

Combinatorial and HTP methodology is a rapidly growing field of electrochemistry,

including new methods for electrosynthesis of small organic compounds,

heterogeneous catalysis and synthesis of conducting solid supports.2,17

The first example of combinatorial electrochemistry was reported by Reddington et

al.30 describing an automated method for preparation of large libraries of metal alloy

catalysts for methanol oxidation. The HTP electrochemical screening of the array

way was based on the optical response of a fluorescent dye due to pH changes caused

by the local proton release during the electroxidation of methanol. Similar optical

methods were reported for HTP electrochemical screening of combinatorial libraries

of electrocatalysts for fuel cells31,32 and Pt-Pb alloy catalysts for enzyme free

amperometric sensors.33 A disadvantage of the HTP optical screening method is

limited sensitivity of the optical response, which can not distinguish small

differences in electrocatalytic current between different members of the array. In

addition, the fluorescent dye can be adsorbed at the electrode surface and affect the

activity of studied catalyst.

This problem was overcome in report by Ward et al.34 for electrochemical HTP

screening of an array of gold electrodes modified with different organothiols, where

fluorescent screen was replaced by more sensitive direct measurement of

electrochemical current. This allowed the authors to distinguish slight variations in

the current recorded for different modifications. Combinatorial electrochemical

screening was accomplished using a computer-automated analysis, in which each

electrode of the library was examined serially in the same electrochemical cell under


7

identical conditions. A disadvantage of this method is that all the electrodes remain

in the solution in contact with the reactants, while only one active electrode is

screened at a time. This might result in inaccurate results obtained for the array of

modified electrodes.

Simultaneous electrochemical screening of all members in the combinatorial array

was achieved by introduction of multichannel potentiostat/galvanostat, which was

initially applied for redox recycling of p-aminophenol - a product of protein and

DNA-enzyme linked sorbent assay with sensitive electrochemical detection.35 In this

method, the same potential was applied to all electrodes followed by automated serial

data acquisition with integrated computer software. Literature examples of HTP

screening using multichannel potentiostat include combinatorial preparation and

screening of lithium batteries36, arrays of Pt-catalysts loaded on carbon for oxygen

reduction, CO electroxidation37 and methanol oxidation.38

In this project, the combinatorial methodology described in Sections 1.1.2 and

1.1.3 using solid-phase synthesis followed by HTP electrochemical screening by a

multichannel potentiostat was applied for discovery of novel electrocatalysts for

biosensor development.


8


9

1.2 Chemically modified electrodes (CMEs)

One of the major aspects of research in electrochemistry within the last 30 years has

been the design of electrode materials in order to provide control over their

interactions with their environment. This has found particular interest in

electroanalytical chemistry, where tailored electrode surfaces with their unique

properties have been used in various practical applications, such as biosensors,

molecular electronics, corrosion protection and energy conversion.39,40

Electrode surfaces functionalised by conducting or semiconducting materials are

defined as chemically modified electrodes (CMEs), where design of these materials

at the surface allows control over the surface performance and properties, such as

selectivity, sensitivity, chemical and electrochemical stability, breadth of potential

window and resistance to fouling.

Immobilised redox active species (mediators) work as catalysts for the

oxidation/reduction of the analyte in solution and allow molecular control over the

kinetics of electron transfer at the electrode/solution interface. The redox reaction of

the solution species occurs through chemical reaction with the redox mediator

followed by electrochemical regeneration of the mediator at the electrode surface

(Scheme 1.2).

Scheme 1.2 General representation of an electrochemical redox reaction catalysed by a mediator.

An effective redox mediator can significantly catalyse a chemical reaction and

electron transfer at low overpotential. The strategy of electrode modification has

been successfully applied in bioelectrochemistry to enable electrochemistry of redox

proteins and enzymes41 and for application in biofuel cells42, biosensors43-45 and

electrosynthesis.46

Mediators can be immobilised at electrode surfaces by covalent or non-covalent bond

formation divided into three general methods:47

• Chemisorption (adsorption) is particularly effective if the mediator contains a

group strongly and irreversibly adsorbing onto the electrode surface. This


10

approach is commonly used in attachment of mediators with thiol functionality

onto gold electrodes yielding monolayer coverage.48 Alternatively, the electrode

surface may be functionalised by strongly adsorbed species with functional

groups, which permit further covalent attachment of the mediator molecules.49

• Immobilisation of polymeric films at the electrode surface can be obtained using

two general methods: coating of a preformed polymer or in situ electrochemical

polymerisation-deposition at the electrode surface. Using these methods,

multilayers of polymer films are created at the electrode surface. These films can

act as a catalyst by themself or be used as a matrix to incorporate mediator

species.

• Covalent functionalisation of the electrode surface with organic molecules

enables control over the surface modification using organic synthesis methods.

Examples of covalently modified electrodes include organosilanes at metal oxide

electrodes such as ruthenium (IV) oxide, Pt/PtO and Si/SiO.39 Formation of

strong Au-C covalent bonds can be obtained by reduction of diazonium salts at

the gold surface yielding stable monolayers and multilayers of the aryl

derivatives at the electrode surface.50,51 Due to the scope of this project, we will

focus on the covalent functionalisation of the GC electrodes.

The use of carbon as an electrode material has the advantages over metal electrodes

in low cost, a wide potential window and electrocatalytic activity for a variety of

redox reactions.52 Different allotropic forms of carbon such as graphite (powdered,

pyrolytic and highly ordered pyrolytic graphite (HOPG) and different carbon

structures such as (vitreous) carbon (GC) and carbon nanotubes (CNT) have been

applied in both analytical and industrial electrochemistry.52 Well-known applications

are examples of commercially available electrochemical blood-glucose strips based

on screen-printed carbon paste electrodes.53

1.2.1 Covalent modification of carbon electrodes

The covalent functionalisation of easily handled and versatile carbon materials

makes them very attractive CMEs with a potentially wide range of practical

applications. The carbon materials are in the form of extended networks of fused

aromatic rings with the terminal surface region of the networks often rich in reactive


11

sites. Hybridised sp2 carbons enable the terminal region to take part in organic

addition reactions or can be easily oxidised to active hydroxyl or carboxylic groups.39

Two main methods of covalent modification of carbon electrodes exist: chemical and

electrochemically assisted. The easily oxidised carbon surface creates reactive

functional groups, including carboxylic acids, ketones and alcohols.47

Electrochemically assisted direct covalent bond formation occurs between aromatic

carbon atoms at the surface and organic molecules in the solution.

1.2.2 Chemical functionalisation of the carbon electrode

Chemically active terminal regions of the carbon surface are easily oxidised to create

a number of functional groups, such as ketones and aldehydes and carboxylic groups.

With such an activated carbon surface, it is possible to covalently attach different

molecules including redox centres or spacer units to create metal complexes.39,54-57

Some oxidation of the carbon surface occurs on exposure of the carbon surface to the

air or during the polishing process. However, in order to increase the density of the

functional groups at the carbon surface, the electrodes are normally pretreated by

heating in air, by oxygen plasma treatment or by oxidation in nitric acid.58,59 The

reactivity of the resulting carboxylic groups can be improved by converting them

into more active acyl chlorides, which can react more rapidly with amines or

alcohols. Thionyl chloride is also used as oxidising reagent, which additionally

increases the surface density of carboxylic groups (Figure 1.4).

Figure 1.4 Functionalisation of the carbon surface by oxidation followed by conversion of the

carboxylic acids into more active acyl chlorides.

Hydroxyl groups at the carbon surface produced by reduction of carboxylic groups

using lithium aluminium hydride as reducing reagent work effectively for the

introduction of organosilanes at the electrode surface (Figure 1.5).60 Although silane

chemistry is very versatile, it is difficult to obtain layers with a well-defined

thickness and they show poor reproducibility and stability. Hydroxyl groups can also

be reacted with cyanuric chloride, which has three reactive sites. This leaves one or

two reactive sites to react with amines, Grignard reagents, alcohols or hydrazines.39


12

Figure 1.5 Chemical functionalisation of the carbon surface by hydroxyl groups followed by

attachment of organosilanes or cyanuric chloride.

1.2.3 Electrochemical functionalisation of carbon electrodes

The formation of a covalent bonds between organic molecules in a solution and sp2

carbon atoms at the surface can proceed via electrochemical heterogeneous reactions

and electrodes prepared by this approach have found a number of analytical

applications. In general, activation of a chemical reaction is initialised by electron

transfer that creates intermediate unstable radicals, which react with carbon atoms at

the electrode surface. However, faradaic efficiency for this type of modification is

low due to the fact that intermediate radicals can dimerise or react with other

reagents in the solution. Electrochemically assisted covalent attachment at the carbon

surface has been successfully accomplished by oxidation of amines61,62 and

arylacetylates,63 reduction of diazonium salt64 and anodization (oxidation) in solution

with alcohols.65

The electrochemical oxidation of carboxylates (Kolbe reaction) is the oldest

known electroorganic reaction.63 The Kolbe reaction is known as decarboxylative

dimerisation and proceeds via a radial reaction mechanism (Figure 1.6). However, if

the oxidation of carboxylate is performed at a carbon electrode, the R groups are

covalently attached at the carbon surface. Firstly, the carboxylate ion transfers an

electron to the electrode surface, yielding carbon dioxide and the R radical (Figure

1.6). The radical R· can then be oxidised to the corresponding carbocation, which

forms a C-C bond with carbon atoms at the surface.65


13

Figure 1.6 Electrochemical functionalisation of electrodes surface by applying Kolbe reaction at

metal and carbon electrodes.

Similarly, amine-functionalised carbon electrodes may be prepared by

electrochemical oxidation of primary aliphatic amines.61,62,65,66 This method is based

on electrochemical generation of primary or secondary amine radicals, which react

with aromatic carbons at the electrode surface to create a covalent bond (Figure 1.7).

X-ray photoelectron studies (XPS) have shown that the extent of the immobilisation

strongly depends on the degree of amine substitution. Porter et al.62 found that

primary amines are much easier to oxidise than secondary amines, while tertiary

amines show no evidence of surface attachment. This might be due to the steric

effect of substituents in the tertiary amines, which limit the access of the amine

radical to surface binding site.

Figure 1.7 Electrochemical functionalisation of carbon electrode by oxidation of primary amines.65

Moreover, the surface attachments with primary/secondary amines or

carboxylic acid, might also consist of the additional active group in their structures,

which enables further introduction of redox centres, linkers and spacers at the

electrode surface using organic synthesis methods. In the case of aliphatic α, ω-

diamines both of the amine groups may be anchored at the carbon surface and create

bridge structures by coupling both of the intermediate radicals of the same diamine

molecule.61 This undesirable effect was overcome by anchoring mono-protected

diamine linkers, where the free amine group was first oxidised at the carbon surface

so that subsequent deprotection of the second amine group leads to an amine

modified carbon surface. These surfaces are analogous of the aminomethyl resins

widely used in solid-phase synthesis.55,57

Examples of this approach include recent work on electrochemical attachment of

mono-Boc protected primary diamines with different lengths and structures

(aliphatic, alkyl and ether) (Scheme 1.3). After electrochemically assisted attachment

of the linkers, an anthraquinone as a model redox centres was covalently attached

using solid-phase synthesis methodology. XPS and electrochemical studies showed

that the length and structure of the linker has an effect on the surface coverage and


14

kinetics of the redox species at the electrode surface. The anthraquinone surface

coverage was found to decrease as the chain length of the diamine linker increased

and faster electrode kinetics were observed for lower coverages and more flexible

linkers.55,57

Scheme 1.3 General scheme of the GC electrodes modified by various diamine linkers and the

anthraquinone redox centre.57

Electrochemical oxidation was also used for the introduction of oxygen functionality

at carbon surfaces.65,67 Formation of ether groups at the carbon surface was possible

by applying high anodic potentials in acidic solution of primary alcohols; this creates

the hydroxyl radical followed by its coupling at the carbon surface. As a result, the

alcohol chain R was covalently bonded at the surface through an ether functional

group (Scheme 1.4). The ether functionalised GC electrode was applied for indirect

detection of calcium ions based on voltammetric response of ferricyanide.67,68

Scheme 1.4 Electrochemical covalent attachment of primary alcohols at the carbon surface.

Derivatisation of carbon electrodes can be achieved by electrochemical reduction of

diazonium salts leading to the formation of a very stable covalent bond between the

aryl groups and sp2 carbon atoms at the surface. This covalent attachment was first

described by Pinson and Savéant,64 who observed an irreversible one-electron

reduction wave in the presence of a diazonium tertafluoroborate salt. The reaction of

wide range of phenyl amines with sodium nitrate at low temperature leads to phenyl

diazonium reagents, usually isolated as tetrafluoroborate salt. Since then,

functionalisation of the carbon electrodes by reduction of diazonium salts has been


15

found a number applications.45,65,69,70-72 The literature on modified electrodes

prepared by reduction of diazonium salts bearing a wide variety of functionality has

been reviewed.65,68,73

Binding of the aryl groups to the carbon surface is likely to be a two-step process.

Firstly, electrochemically assisted one-electron reduction of the diazonium salt

generates an aryl radical and molecular nitrogen. This is followed by coupling of the

radical to sp2 carbon atoms at the electrode surface (Scheme 1.5).74 Covalent

anchoring of the diazonium salt to the electrode surface is favoured due its ability to

adsorb on the surface and the relatively positive potential for diazonium reduction of

about – 40 mV vs. SCE, which prevents simultaneous reduction of the aryl

radical.64,65

Scheme 1.5 Electrochemical functionalisation of carbon electrodes by reduction of a diazonium salt.

The very active phenyl radicals may react with phenyl groups already attached on the

surface yielding multilayers. This occurs under certain conditions, particularly at

high diazonium salt concentration and long reduction times and allows deposition of

the multilayers of the aryl compounds at desired thickness up to 25 nm.65,74,75

Introduction of aryl derivatives bearing active functional groups enables further

modification of the electrode surface by attachment of additional redox centres,

spacers or linkers using solid-phase organic synthesis methods, as shown in recent

work by Bartlett and Kilburn.55 In this work, the diazonium salt was attached at the

GC surface, followed by further functionalisation of the electrode using solid-phase

methodology (Scheme 1.6). An electrochemical reduction of 4-(N-Boc-

aminomethyl)benzene diazonium tetrafluoroborate salt allowed for formation of a

monolayer of benzylamine at the GC surface due to the fact that the bulky Boc group

prevents the formation of multilayers of benzyl groups. After removal of Boc-

protecting group, an anthraquinone carboxylic acid (AQ) redox centre was attached

using solid-phase conditions. The modified electrodes were electrochemically

characterised and show good long-term stability. In comparison to the analogous

modification by mono-Boc protected diamine linkers, GC electrodes modified by the

diazonium salt show slower electron transfer and lower surface coverage after


16

attachment of the AQ redox centre. This strategy, which combines electrochemical

and solid-phase synthetic methodologies, constitutes a versatile and flexible

approach for functionalisation of different carbon materials.

Scheme 1.6 Covalent functioanlisation of the GC electrodes by 4-(N-Boc-aminomethyl)benzene

diazonium tetrafluoroborate salt linker and AQ redox centre; a) from 0.6 to −1 V vs. Ag/AgCl in 0.1

M TBATFB in MeCN; b) 4.0 M HCl in dioxane; c) Anthraquinone-2-carboxylic acid (AQ), HBTU,

DIEA, DMF, R.T., 16 h.55


17

1.3 NADH co-enzyme and its electrochemical oxidation

Redox enzymes (oxidoreductases) exist to catalyse both electron transfer

reactions and the transfer of atoms or small groups of atoms, to or from a range of

substrates in a large number of biological processes. Some of these require small

non-proteinaceous molecules, called co-enzymes in order to be active. The co-

enzyme acts as an acceptor or donor of a small group of atoms or electrons and

provides the driving force for the redox reaction of the substrate.43

An important group of redox enzymes are dehydrogenases, responsible for the

transfer of hydrogen atoms and electrons during biological reactions and pathways in

living organisms. Over 300 dehydrogenases are known, each of them specific for one

substrate or group of substrates.43 Most of the characterised dehydrogenases are

dependent on soluble pyridine nucleotide cofactors, which occur in two biologically

active forms of �-NAD (nicotinamide adenine dinucleotide) and its phosphorylated

version, bearing a 2’-phosphate at the ribose ring �-NADP (nicotinamide adenine

dinucleotide phosphate) (Figure 1.9). These coenzymes work as an acceptor/donor of

a hydride ion (H−) to/from a substrate during oxidoreductase-catalysed redox

reactions. The NAD(P) coenzyme reacts with the substrate in 1:1 stoichiometric

reactions. Changes of the NAD(P) concentration in solution during enzymatic

reaction can be monitored by UV spectroscopy since NADH has strong adsorption at

wavelength of 340 nm.76

In general, the active site of a redox enzyme contains complexing groups, which

binds with the substrate and co-enzyme NAD(P)+/NAD(P)H to bring them into

correct relative orientation (Figure 1.8). The NAD(P)+/NAD(P)H co-enzyme binds

first at the active site to give a haloenzyme, followed by binding of a substrate. The

redox reaction between enzyme and the substrate transfers atoms and electrons

to/from the electron acceptor, which in this case is the co-enzyme, followed by

dissociation of the product and coenzyme from the active site.


18

Figure 1.8 The general mechanism of an enzymatic reaction catalysed by NAD(P)+/NAD(P)H co-

enzyme.

In the case of the dehydrogenases, the enzymatically active NAD(P)+ works as an

acceptor of two electrons and one proton (hydride transfer H−) during substrate

oxidation to obtain NAD(P)H.76,77 The redox active part of this coenzyme is the

nicotinamide ring, which exists as pyridinium salt for oxidised NAD(P)+ and as the

1,4-dihydropyridine form for reduced NAD(P)H (Figure 1.9).

N

R

O

NH2

H H

OH OH

O

O

N

P

O

OH

P

O

OH

OOO

OH

O

N

N

N

1,4-NAD(P)H

N+

R

O

NH2+2e-, +H+, (or H-)

R=

X= H (NAD+) ,PO32- (NADP+)

-2 e-, H+

X

NAD(P)+

Figure 1.9 Structures of NAD and NADP and their reversible redox reaction by hydride ion transfer

between oxidised form of NAD(P)+ and reduced form of NAD(P)H carried out at C-4 of the

enzymatically active nicotinamide ring; substituent R represents enzymatically inactive linkage

consists of ribose, adenine base and phosphate groups and substituent X at one of the ribose rings is H

for NAD and a phosphate group PO32- for NADP.

The rest of the molecule including the adenine rings, ribose and phosphate groups are

responsible for specificity and affinity of binding with the active site of the

enzyme.78 During the enzymatic redox reaction, hydride transfer occurs with one of

the C-4 hydrogen atoms at the nicotinamide ring. Dehydrogenase enzymes are each


19

specific for one of the enantiotopic C-4 protons of NAD(P)+/NAD(P)H: some of

them react with the 4R hydrogen and the others with the 4S hydrogen (Figure 1.10).

The stereospecificity of the dehydrogenase enzymes can be determined by well-

established deuterium labelled experiments.76,79,80 Stereospecificity of the

dehydrogenase enzymes is due to the orientation of the NAD+/NADH in the enzyme

active site.76

Figure 1.10 Stereospecific hydride transfer during redox reaction between NAD(P)+ and NAD(P)H.

In solution, NAD(P)+ exists in a folded conformation where parallel-ring stacking of

the adenine and pyridine molecules creates an intramolecular complex with an

average inter-ring distance of 0.52 nm and the nicotinamide ring si-side facing the

adenine and confirmed by spectroscopic analysis 81 and molecular dynamics

simulation.78,82

The unfolded, extended structure of the coenzyme is observed in the case of

NAD(P)+/NAD(P)H coordinated with proteins in the enzyme active site.83 For most

dehydrogenase enzymes, extended NAD+ binds to a pair of protein structural

domains, known as the Rossmann fold. The Rossmann fold typically consists of six

parallel amino acid β-strands and four associated α-helices, with the adenine and

nicotinamide rings interact with a pair of β-α-β motifs.84,85 For dehydrogenase

enzymes, it is well known that the domain contains two sets of β-α-β-α−β units

joined across a two-fold axis such that the first strands of each unit are adjacent. The

strand order is thus 6 5 4 1 2 3 (Figure 1.11).86,87 The only adjacent strands, which

have helical connections in opposite directions are 1 and 4. They create cervices,

where the phosphate groups of the coenzyme bind to the corresponding carboxy ends

of these two strands. The long loop (cross-over) between strands 3 and 4 creates a

cavity that participates in the binding of the adenine ring of the coenzyme.


20

Figure 1.11 Schematic diagrams illustrating the Rossmann folding of β-strands and α-helices in the

NAD-binding protein domain of dehydrogenase enzyme.87

Despite their very similar structures, identical thermodynamic properties and

reaction mechanism, �-NADP and �-NAD coenzymes participate in different

biological processes within the living cell.77 �-NADP works with enzymes that

catalyse anabolic reactions supplying electrons needed to synthesise energy-rich

biological molecules. �-NAD is found in mitochondria where it is essential for the

conversion of ADP to ATP through the oxidation of foodstuffs (catabolic function).

ATP molecules are used elsewhere in the cell to provide the energy for

thermodynamically unfavourable reactions required to build essential cell

components. Most nicotinamide-dependent dehydrogenase enzymes are specific for

NAD(P)+/NAD(P)H or the form (NAD+/NADH) of the co-enzyme.

NAD(P)-dependent dehydrogenase enzymes catalyse oxidation of their substrates

according to the general reaction scheme:

k+ +2 k

SH + NAD(P) S + NAD(P)H + Hf

b

→←

where S and SH2 are the oxidised and reduced forms of the substrate and kf and kb are

the rate constants for the forward and backward reactions. This general enzymatic

reaction in the presence of NAD+ is modified, depending on the structure of the

substrate and classification of those enzymatic reactions has been reported in the

literature.77,88 NAD(P)+ coenzymes are most commonly used in enzymatic oxidation

of alcohols to carbonyls, including alcohol dehydrogenases, which catalyse oxidation

of ethanol to acetaldehyde.

k+ +

kethanol+NAD acetaldehyde+NADH+H

f

b

→←


21

The biological importance and the redox properties of NAD(P)+/NAD(P)H

coenzyme have attracted significant interest for its application in enzymatic synthesis

catalysed by NAD(P)H-dependent enzymes and amperometric biosensors. The

enzyme-catalysed synthesis of different compounds has found practical application

in pharmacology, biotechnology, food additives, perfumes, pesticides and

insecticides.89 Because of the very high price of NAD(P)H and the fact that it needs

to be used in stoichiometric quantities during enzymatic reaction, its use in industry

is limited and needs to be justified by the high price of the final product. Hence, there

is significant interest in developing in situ NADH regeneration methods to re-use the

initial amount of the active NAD(P)H. One of the most popular regeneration methods

include reduction of NAD+ to NADH by formate catalysed by formate

dehydrogenase (FDH) enzyme. Electrochemical methods of the NAD(P) would

remain the best means of regeneration because these do not need an additional

donor/acceptor of the electrons and no side products are produced. Similarly, the

redox properties of the NAD(P) co-enzyme have found extensive application in

design of the amperometric biosensors. However, electrochemical behaviour of the

NAD(P) at a bare electrode is complex and results in number of site reactions,

therefore bioelectrocatalysis, new electrochemical mediators and modifications to the

electrode surface have attracted a lot of attention.

1.3.1 Electrochemistry of NADH/NAD+

The development of electrochemical biosensors based on NADH-dependent

dehydrogenases is an attractive goal for many researches mainly due to the large

field of potential substrates to analyse and their diversity in commercial applications.

The electrochemistry of the redox couple NAD+/NADH has been extensively

studied. The experimentally obtained formal redox potential (E°') of the

NAD+/NADH couple is −0.315 V vs. NHE and −0.560 vs. SCE at pH 7.90,91 The

NAD(P) dependent biosensors found practical application due to the fact that the

redox process of NADH(P) is more favoured than most of the substrates catalysed by

the NAD(P) coenzyme as value of E°' for NAD(P) is lower than most of the

substrates. In particularly, biosensors based on oxidation of NADH to enzymatically

active NAD+ were considered for more detailed studies and development. Biosensors

based on the reduction of NAD+ would be impractical due to a series of interfering


22

side reactions occurring at more negative potentials, such as oxygen reduction, which

would disturb the reliability of the biosensor.

Initial studies on NAD redox behaviour at bare electrodes indicated that the

electrochemistry of the NAD+/NADH is irreversible with an anodic wave observed at

high overpotentials.92 The choice of electrode material has a significant effect on the

anodic potential with values of 0.4, 0.7, 1 V vs. SCE for carbon, platinum and gold

electrodes, respectively.93,94 It has been established that the oxidation of the NADH

in aqueous solution occurs via a two-electron process77,92,95

+ + -NADH NAD + H + 2e→

Electrochemical oxidation of the NADH at bare electrodes depends on the electrode

material, NADH concentration and electrode pre-treatment. Quantitative recovery of

NAD+ was obtained after electrochemical oxidation at Pt electrodes at low NADH

concentrations (NADH < 0.1 mM).92,94 Control experiments for the pre-adsorbed

NADH at Pt and Au surfaces showed that NADH is oxidised at the electrode surface

and undergoes further oxidation processes leading to unspecified products, which

might result in fouling of the surface.93

Studies on the oxidation of NADH at the carbon electrodes showed that the oxidation

product NAD+ adsorbs strongly at the surface and acts as an inhibitor for further

NADH oxidation.93,96 This results in shifts to anodic potential with changing NADH

concentration. A stable anodic potential was observed, when NAD+ was present in

excess in the solution or when the electrode was pre-treated with NAD+ prior to the

electrochemical experiment.

1.3.2 Mechanism of direct electrochemical NADH oxidation

Reviews by Bartlett and Gorton provide a summary of the mechanistic studies on the

electrochemical oxidation of NADH at bare electrodes.77,97 In general, investigation

on the mechanistic aspects of the NADH oxidation was based on discussion whether

the NADH electrochemical reaction occurs as a single two-electron step or involves

consecutive one-electron transfer steps. Oxidation of NADH model compounds at Pt

electrodes in aprotic solvent and without the presence of base gave strong indication

that reaction occurs in two one-electron steps and is highly irreversible.95 A number

of further experimental studies at NAD+ coated electrodes suggested that

electrochemical oxidation of NADH proceeds according to an ECE mechanism


23

+- -

+

-H-e -e•+ • +

+HNADH NADH NAD NAD→→ →←

The first step is an irreversible, potential-determining electron transfer, yielding a

cation radical NADH+·. The high activation energy required for the first step might

be due to the fact that 1,4-NADH must undergo rearrangement before losing the

electron (Figure 1.12).98 The amide group can take part in the oxidation process since

only 1,4-NADH is enzymatically active.

The enol resonance form might also participate in the electron transfer as was

observed during oxidation of the NADH by orthoquinones.98,99 The second step

involves deprotonation of NADH+· yielding the neutral radical NAD·. This step is

also considered as irreversible since the presence of the radical at the pyridine

aromatic ring is more stable than the cation radical.98,99 The deprotonation is

relatively fast with a calculated rate constant kH+ > 106 s−1 measured for NADH

oxidation at GC electrodes coated with NAD+. NAD· radical can be rapidly

dimerised as has been identified during reduction of NAD+, where formation of

dimer (NAD)2 from coupling of two radicals NAD· was observed at about −1.1 V vs.

SCE.100 The dimer is oxidised at −0.4 V vs. SCE and yields NAD+. Therefore,

oxidation of NAD· to NAD+ in the final step of oxidation of NADH would be

sufficiently rapid to outrun the dimerisation and practically no dimers would be

produced.77


24

Figure 1.12 Proposed reaction pathway for the electrochemical oxidation of NADH according to the

ECE mechanism; kH+ is a deprotonation rate constant.

In addition, the neutral radical NAD· might exchange an electron with the more

reactive cation radical NADH+ during a chemical disproportionation reaction (DISP),

leading to NADH and NAD+

•+ • +NADH + NAD NADH + NAD⇔

Studies by Blankespoor and Miller96 observed that 95% of the oxidation product is

formed through the ECE pathway. The predominance of the ECE pathway might be

explained by the fact that significantly rapid deprotonation of NADH+· creates NAD·

radicals at a short distance from the electrode surface. The NAD· generated close to

the surface having relatively high potential, can easily lose an electron before it

diffuses away from the electrode surface. Hence, electrochemical electron transfer

(ECE) is favourable over the homogenous electron transfer (DISP), when the rate

constant kH+ is relatively high.

In the case of electrodes, which are not covered with adsorbed NAD+, the

NADH oxidation might be mediated to some extent by oxygen adsorbed at the Au

and Pt eletrodes (eg. OH·

ads/H2O and Oads/OH·

ads) or by organic functional groups

created during oxidation at carbon surface (e.g. quinone, semiquinone and

hydroquionone systems).77,93,101

The electron-transfer path involves electron exchange between energy levels located

at the surface oxygen atoms and the electrode, coupled with the electron exchange


25

between energy levels of the surface oxygen atoms and the solution species (Figure

1.13).

Figure 1.13 General scheme of the possible electron transfer pathway during NADH oxidation

mediated by surface oxygen redox systems; a) mediated by hydroxy groups; b) mediated by molecular

oxygen.77,93,101

The proton transfer pathway includes transfer of the proton from the C-4 of the

nicotinamide ring to the proton acceptor, such as a water molecule, and possible

intermediate formation of a bond to the surface oxygen atoms (Figure 1.14).

Figure 1.14 Possible proton transfer mechanism during oxidation of NADH involving movement of a

proton from C-4 of the nicotinamide ring to a proton acceptor such as water molecule.77,93,101

The existence of functional oxides at the carbon surface is well established and is

known as one of the methods of carbon surface modification (Section 1.2.2).

However the type of the oxygen functional groups and their quantity at the electrode

surface varies between different carbon materials and oxidation method.77 Studies on

oxidised carbon electrodes suggested that possible oxygen redox mediators at the

surface might decrease the anodic potential for NADH oxidation causing the

oxidation reaction to become kinetically more rapid. However, this electrocatalytic

effect was found to be of low stability, which might be due to surface blocking by the

oxidation product NAD+.77,94


26

1.3.3 NADH oxidation at modified electrodes

In order to design effective biosensors based on the electrochemical oxidation

of NADH, pre-treatment of the electrode surface is required to overcome problems of

the high overpotential and ECE mechanism, resulting in fouling of the electrode

surface. During initial studies on oxidised Pt and carbon electrodes a decrease of the

overpotential for NADH oxidation to approximate 0.2 V vs. SCE was observed. This

suggested that presence of quinone and hydroxyl groups at the electrode surface act

as mediators for electron transfer between the electrode and NADH molecules.94,99

Since then, a large number of the redox systems have been investigated as

electrocatalysts for NADH oxidation, including homogenous electron-transfer

mediators, which are known to successfully catalyse the NADH oxidation.97

However, the disadvantage of using soluble mediators is the possibility that they can

diffuse away from the surface and no longer be available to catalyse the electrode

reaction. Hence, most of the mediators applied for NADH oxidation are immobilised

at or very close to the electrode surface to create CMEs.

The perfect mediator has to meet a number of criteria:

I. Acceptor of the hydride ion

An essential feature of the effective NADH mediators is their ability to act as two

electrons (2e−), one proton (H+) acceptors during oxidation of NADH to produce

enzymatically active NAD+. The reaction between NADH molecules and the

mediator occurs via formation of an intermediate charge transfer complex with the

NADH. Efficient electrocatalyst immobilised at the electrode surface reacts rapidly

with the NADH at low overpotential. This means that the oxidation of NADH to

NAD+ should proceed via single hydride transfer H− rather than sequential one

electron transfers since the radical intermediates for NADH oxidation are unstable

and undergo side reactions. After oxidation of the NADH, the reduced form of the

mediator must be regenerated at the electrode surface to the reoxidised form. Bartlett

et al.102 proposed a general reaction mechanism, where the mediator reoxidation

proceeds via sequential one electron and proton steps and may vary depending on the

mediator and conditions (Figure 1.15).


27

Figure 1.15 The catalytic cycle for the oxidation of NADH by the mediator M immobilised at the

electrode surface.43,102

II. Low overpotential

The perfect mediating system should be able to reduce the overpotential down to

optimal range of −0.1-0.1 V to avoid an oxidation of substrates such as ascorbate,

urate and acetaminophen and avoid the potential region for reduction of molecular

oxygen and where the potential of zero charge results in low a background current

and noise for most of the electrode materials. For design of the amperometric

biosensors, low redox potential of the mediator also prevents oxidation of other

species in the biological sample.

III. Fast reaction kinetics

The mediator should be able to react rapidly with the NADH coenzyme followed by

fast electron transfer during electrochemical regeneration of the mediator at the

electrode surface. Ideal second order rate constants for the reaction between the

NADH and the mediator should be above 106-107 M−1 s−1, approaching a diffusion

controlled reaction. In the case of dehydrogenase enzymes, the high rate constant

between NADH and mediator kobs is particularly important due to the fact that

reaction between substrate and coenzyme is thermodynamically favourable and

reaches its equilibrium rapidly. As a result, the rate kobs needs to be higher than the

rate of the back reaction kb in order to increase the rate of product formation kf

(Figure 1.16). If kobs < kf, the electrochemical response will be mostly dependent on

the product concentration.77


28

Figure 1.16 Enzymatic reaction of the substrate S catalysed by NAD+ followed by electrochemical

reoxidation of the NADH by the mediator M at the rate constant kobs, which is in kinetic competition

with the rate constant kb of the enzyme catalysed back reaction of NADH with the product P.

IV. Long-term stability (at least weeks or months)

This includes irreversible attachment of the mediator at the electrode surface

followed by its electrochemical and chemical stability (hydrolysis, chemical

oxidation or photo-decomposition). In addition, the mediating systems should not

take part in the radical site reactions with NADH, should be selective towards

NADH and show well-defined stoichiometry.77

A first attempt at a CME for NADH oxidation was reported by Kuwana and

Tse99 where they covalently immobilised primary amine-containing o-quinones at a

carbon surface activated with cyanuric-chloride. These redox systems at the surface

showed an anodic peak in the presence of NADH at around 0.2 V vs. SCE. However,

a substantial decrease in the stability of the CMEs was observed during NADH

electroxidation. Further investigation for o-quinones incorporated into a larger

aromatic derivative 4-(2-(1-pyrenyl)vinyl)catechol), reported by Kuwana et al.,103

indicated long term stability of the mediator in its reduced form. When the strongly

adsorbed mediator was kept at a potential close to its formal redox potential, site

reactions occurred caused by reaction of the o-quinone groups with catechol.

Additionally, a deactivation reaction of the mediator in its reduced form was

suggested, where they were reacting with the NADH+· radical to give an inactive

compound leading to poisoning of the electrode surface (Figure 1.17).

Figure 1.17 Possible deactivation mechanism for the adsorbed o-quinone mediator and NADH+·

radical during electrochemical oxidation of the NADH.77,103


29

Since the first attempts at preparation of CMEs for electroxidation of NADH, a

significant amount of work has been reported on the development of the CMEs and

is summarised in a number of reviews.77,97,104,105 In general, two main development

routes were identified during evolution of the CMEs. The first, involved searching

for different methods for the immobilisation of mediators at the electrode, discussed

in more detail in Section 1.2. The second route involved the development of new

mediator structures, other than o-quinones, which could effectively catalyse NADH

oxidation at low overpotential and have long-term stability.

In general, redox systems contain the structural features depicted in Figure 1.18

are found as effective 2e−, H+ acceptors and to successfully catalyse NADH

oxidation at low overpotential.43 The essential features of the effective NADH

mediator are the presence of the hydride accepting group and the ability to delocalise

the charge within the mediator (Figure 1.18).

Figure 1.18 Proposed mechanism of the NADH oxidation catalysed by redox mediator, X is a group

working as a hydride acceptor; Y is an electron deficient group.106,107

The most commonly suggested model for the reaction between the mediator and

NADH is a two-step reaction mechanism,108 analogous to the Michaelis-Menten

enzyme kinetic model, described in Section 1.5.

Effective NADH mediators containing the necessary conjugated structures can be

differentiated into a few general classes depending on the immobilisation method and

the structure of the mediator:

I. Quinones

Since the initial study by Kuwana et al.,99 a number of workers have reported CMEs

with quinone and dihydroxy functionalities. Examples includes ortho- and para-

dihydroxybenzaldehyde derivatives electropolymerised or adsorbed at the electrode

surface, which were found to effectively catalyse NADH oxidation at low

overpotential and exhibit a good stability during electrochemical screening.109


30

Recently, a novel HTP methodology for the functionalisation of GC electrodes was

reported with use of the dihydroxy derivatives as mediators. The GC electrodes,

covalently modified by two different diamine linkers (Section 1.2.3) were coupled by

3,4-dihydroxybenzaldehyde derivatives in a combinatorial way, followed by their

electrochemical screening towards NADH oxidation.56 This novel approach allowed

for rapid evaluation of various dihydroxybenzene mediators towards electrocatalytic

NADH oxidation with the highest catalytic activity obtained for electrodes modified

with p-benzylamine linker and 3,4,5-trihydroxybenzene. Disadvantage of the

dihydroxybenzene mediators is their low electrochemical stability.

II. Redox dyes

Commercially available phenoxazines and phenothiazines redox dyes such as

meldola blue, nile blue or methylene blue effectively catalyse NADH oxidation at

potential about −0.1 V vs. SCE. In general, this type of mediator shows relatively

low chemical and physical stability. In order to overcome this problem, the redox

dyes were derivatised in position 3 or 7 with amine functionalities, aromatic

aldehydes or acid chlorides to donate beneficial new properties to the original dye,

such as resistance to the pH changes or higher adsorption stability.

Figure 1.19 Examples of common redox dye mediators applied as effective catalysts for NADH

oxidation.

The redox dyes were also incorporated into carbon paste,43 electropolymerised

(poly(nile blue)110 or poly(methylene blue)111) at the electrode or covalently attached

to a compound strongly adsorbed at the surface (monolayers of various redox dyes

attached at the gold surface through self-assembly cysteamine or 3,3'-

dithiobis(succineimidylpropionate) in order to reduce the desorption problem and

catalyse NADH oxidation at desired formal potential about −0.1 V.112,113

III. Conducting polymers

Conducting polymers such as poly(aniline) doped with poly(anions) such as

poly(vinylsulphonate), poly(styrenesulphonate) or poly(acrylate) show catalytic

activity for NADH oxidation at potentials around 0.1 V vs. SCE.102,106,114 The


31

electrodeposited polymers can act as a host matrix for the mediator, which can be

incorporated into the polymer film as a counter ion during the polymerisation process

or covalently attached to the monomer and then polymerised at the electrode.

Examples of the mediators incorporated into the polymer matrices include pyrrole

substituted by mediators with quinone functionalities115 or the redox dye methylene

blue incorporated into poly(pyrrole) film, which oxidises NADH at about 0.1 V vs.

SCE.116

IV. Metal Complexes

Metal complexes with redox active ligands constitute an interesting group of

mediators towards NADH oxidation. They were normally electrodeposited or

electropolymerised at the electrode surface. Examples include polyvinylimidazole

(PVI) coordinated to an osmium metal ion and deposited at carbon fibre electrodes to

catalyse NADH between 0.1 and 0.5 V vs. Ag/AgCl.117 Other examples include

various catechol-pendant metal complexes (Co, Cr, Os, Ru, Fe, Ni) with terpyridine

electrodeposited onto GC electrodes which were found to effectively reduce the

NADH oxidation potential to about 0.3 V vs. SCE.118 In these cases, the NADH

oxidation is governed by the redox process of the metal ion and have found practical

application in peroxidases based biosensors.119 However, this method has a

disadvantage of necessity of addition of soluble quinoid compound that in a first step

oxidise NADH to NAD+ and then is reoxidised by molecular oxygen to produce

hydrogen peroxide, which can react with peroxidase and osmium complex

immobilised at the surface and generate the response current.105

An interesting group of the mediators for NADH oxidation are metal complexes with

redox active 1,10-phenanthroline-5,6-dione ligands, their more detailed

characterisation is presented in the following Section.

V. Other redox systems

CMEs based on conducting organic salts such as N-methylphenazinium

tetracyanoquinodimethane (NMP-TCNQ) (Figure 1.20) work effectively as catalysts

for NADH oxidation at low overpotentials around 0.2 V vs. SCE and with relatively

high reaction rates.120


32

Figure 1.20 Structures of the organic conducting salts N-methylphenazinium

tetracyanoquinodimethane (NMP-TCNQ) as effective mediators for NADH oxidation.

Organic compounds containing nitro functional groups such as nitro-fluorenone

derivatives121,122 or dithio-bis(2-nitropyridine)123 can be electrochemically reduced to

nitroso/hydroxylamine, which can then catalyse the NADH oxidation at potential

around −0.15 V vs. SCE as shown in Figure 1.21.

Figure 1.21 a) Examples of the mediators consisting of the nitro functional groups; b) electrochemical

formation of the nitrosyl catalytic active form.122


33

1.4 Metal complexes with 1,10-phenanthroline-5,6-dione as

electrocatalysts for NADH oxidation.

An interesting group of mediators towards NADH oxidation are metal complexes

with the redox active phenathroline-5,6-dione (phendione) bidentate ligand, where

the quinone functionalities act as hydride acceptors during the NADH

electroxidation.

The structure of the non-coordinated phendione ligand contains of two

electrochemically active sites: the o-quinone and imine functionalities. Some of

electrochemical properties of the phendione and its derivatives have been reported in

the literature due to their interest in synthetic and biological applications.124,125 Study

on the electrochemical behaviour of the quinone functionalities suggested that their

electrochemistry strongly depends on the reaction conditions, such as pH and

solvent.126

Cyclic voltammetry of phendione in aprotic solvents shows two reversible-one

electron waves at –0.45 and –1.25 V vs. SSCE, representing formation of the stable

intermediate anion radical (semiquinone) and dianion, respectively (Figure 1.22).

Figure 1.22 a) The reaction pathway of the phendione redox process in aprotic solvent; a) Cyclic

voltammogram for 1.2 mM solution of the phen-dione in MeCN containing 0.1 M TBAP at graphite

electrode.127

In aqueous solvents, the reduction of the quinone to the 5,6-dihydroxy occurs as a

two-electron, two-proton process so that only a single, pH-dependent electrochemical

wave is observed.

N

N

O

O

N

N

O

-O

+1e-N

N

-O

-O

+1e-


34

Figure 1.23 a) General scheme of the reversible electrochemical redox reaction of the phendione

ligand in aqueous solvent; b) Cyclic voltammograms for 1.1 mM solution of phendione in aqueous

solution at pH 2.85 (curve B) and pH 6.8 (curve C) at a Pt electrode and sweep rate 0.2 V s−1.127

Abruña et al. reported a shift of 63 mV/pH unit in aqueous buffer solution, close to

expected value of 59 mV/pH unit for two-electron, two-proton process.88 At pH > 5,

an irreversible reduction peak was observed, which might result from intramolecular

complexing of the 5,6-dihydroxy groups and unprotonated imine functionalities

through hydrogen bonding (Figure 1.24).128,129

N

NO

OH

HN

NOH

OH

Figure 1.24 Formation of the hydrogen bridge between 5,6-dihydroxy and imine groups of the

phendione species as proposed to explain of the irreversible cyclic voltammograms observed at

pH>5.126,129

The problem of the irreversibility at pH > 5 was overcome by coordinating the imine

groups with a transition metal ion to obtain metal complexes. The phendione forms

stable metal complexes with a wide range of the transition metal ions, which

decreases the electron density within the three ring system of the ligand and shifts

redox potential of the quinone groups to more positive values. As a result, such metal

complexes potentially allow for the variation of the redox properties of the

phendione quinone groups including the tuning of the potential through pH

changes.127

Examples of the synthesis of metal complexes with phendione ligands and

their application for NADH oxidation are reported in the literature. Abruña et al.127

presented the synthesis and electrochemical characterisation of Ru, Co, Fe and Os

N

N

O

O

+2e-, 2H+N

N

HO

HO


35

complexes with the phendione in the general forms of [M(phendione)3]2+ (Figure

1.25a), [M(phendione)2(bpy)]2+ and [M(phendione)(bpy)2]2+, where bpy is a 2,2’-

bipyridine ligand. Cyclic voltammograms of the metal complexes in aprotic solvents

showed two waves corresponding to the anion radical and dianion of the quinone

group at potentials more positive than for the free phendione ligand. A one electron

wave was observed at relatively positive potential from 0.65 V vs. SSCE for

[Co(phendione)3]2+ to about 1.4 V vs. SSCE for various tris ruthenium complexes

bearing from one to three phendione ligands coordinated to the ruthenium metal

centre (Figure 1.25b).

Figure 1.25 a) Structure of the metal complex of general form of [M(phendione)3]2+; b) Typical cyclic

voltammogram at a Pt electrode at scan rate of 20 mV s−1 for 0.5 M [Fe(phendione)3]2+ in aprotic

solvent with 0.1 M TBAP, reported by Abruña et al.127

The osmium complexes were adsorbed at the electrode surface and showed

significant catalytic activity towards NADH oxidation which decayed with time due

to desorption of the complex. The ruthenium complexes in homogenous solution

were evaluated as electrocatalysts for NADH oxidation and showed significant

catalytic currents at about 0.05 V vs. SSCE. In addition, electrocatalytic activity of

the [Ru(phendione)3]2+ complex in solution was evaluated in the presence of malate

and lactate dehydrogenase enzymes and it was confirmed that this complex works as

an effective electrocatalyst for NAD+ regeneration after an enzyme turnover.130

Other examples include osmium and ruthenium complexes with phendione

ligand(s) mixed with carbon paste electrodes (CPE), which provide stable and

reversible systems for electroxidation of NADH at low overpotential at around 0.1 V

and 0.05 vs. Ag/AgCl for the osmium and ruthenium complexes, respectively.131,132

The modified CPEs electrodes were applied for reagentless ethanol biosensors,

where Ru and Re complexes with phendione, NADH and alcohol dehydrogenase


36

were incorporated into the carbon paste and worked effectively as ethanol

biosensors.133

The metal complexes of general structure [M(phendione)3]2+ (M= Fe, Ru, Co,

Cr, Ni) or [Ru(v-bpy)2(phendione)](PF6)2 (v-bpy is 4-vinyl-2,2’-bipyridine ligand)

and [Re(phendione)(CO)3]2+ were electrochemically deposited or electropolymerised

at the GC surface and worked as effective mediators for NADH oxidation at

potentials around 0 V vs. SCE.116 The highest catalytic activities during NADH

oxidation were observed for the electrodes with electrodeposited [Fe(phendione)3]2+.

This complex also exhibits the fastest kinetics, with a calculated second order rate

constant of 5.6 × 103 M−1 s−1, probably due to the larger number of the phendione

ligands present in comparison to the Re or Ru complexes.

Popescu et al.134 reported a more detailed study on the electrochemical behaviour of

osmium complexes with the phendione ligand adsorbed at graphite electrodes and its

electrocatalytic activity towards NADH oxidation. As a result, they confirm that the

redox process of the phendione is a two electron, two proton reaction with an

estimated rate constant for the heterogeneous electron transfer between phendione

and the electrode surface of about 20 s−1. In addition, this mediator at the surface

exhibits a significant and persistent electrocatalytic activity towards NADH

oxidation with calculated values of the second order rate constant about 1.9 × 103

M−1 s−1.


37

1.5 Enzyme kinetics - the Michaelis Menten approximation

In 1905 Henri and Brown reported studies on reactions catalysed by enzymes in

which they suggested that the fundamentals accepted at that time for chemical

catalysis were not applicable for enzymatic reactions.76,77,135 Further work by

Michaelis and Menten gave an insight into the mechanism of enzymatic reactions.

Based on studies of conversion of sucrose catalysed by invertase or saccarase

enzymes, they found that the velocity of the enzymatic reaction is dependent on the

substrate concentration S, resembling hyperbolic curve (Figure 1.26). They

suggested a mechanism, where enzyme E and substrate S first form an intermediate

complex ES and this subsequently breaksdown, giving a product P and the free

enzyme

1 2

-1

S+E SE P+Ek k

k

→ →←

where k1, k-1 and k2 represent the rate constants for the individual steps. The

theoretical model proposed by Michaelis and Menten was based on the following

general assumptions:

• Formation of the ES complex is reversible and reaction between E and S

remains in equilibrium. Any effects interfering with this equilibrium are

negligible.

• Forward and backward reactions of the intermediate complex ES reaches the

equilibrium rapidly and concentration of ES remains constant on the time-scale

of the enzymatic reaction (steady-state assumption).

• The substrate S is assumed to be in a great excess over the enzyme E (S >> E),

therefore the concentration of free substrate S can be taken as an equal to its

initial concentration at every point during the reaction.

• Formation of the product P is a rate determining step (k2 << k1 and k2 << k-1)

and its initial velocity (rate) V0 equals

0 2

d[P][ES]

dV k

t= = 1.1

The total concentration of enzyme [E0] is a sum of the concentration of free enzyme

[E] and the concentration of enzyme bounded to the substrate in [ES] complex


38

[ ] [ ] [ ]0E = E + ES 1.2

The concentration of the intermediate ES is difficult to determine experimentally,

therefore a different expression needs to be used in order to calculate values of the

initial velocity V0 of the enzymatic reaction.

When, the reversible formation of ES complex stays in equilibrium, the rate of

formation of ES is equal to the rate of consumption of ES:

Velocity of the ES formation:

[ ][ ] [ ]( )[ ]1 0

d ESE - ES S

dV k

t= =

Velocity of the ES breakdown

[ ][ ] [ ]1 2

d ESES ES

dV k k

t−= = +

1.3

1.4

Mathematically, this assumption means:

[ ] [ ]( )[ ] [ ] [ ]1 0 1 2E - ES S ES ESk k k−= + 1.5

And leads to an expression for the concentration of the intermediate complex [ES]

[ ][ ][ ]

[ ]1 0

1 2 1

E SES

S

k

k k k−

=+ +

1.6

Combining this relationship with the expression for velocity of product formation:

[ ][ ]

[ ][ ]

[ ]

2 00 2

1 2

1

d P E SES

d S

kV k

k kt

k

−

= = =+

+

1.7

The equation for the initial velocity of product formation is obtained, known as

Michaelis-Menten equation:

[ ][ ]0 max

M

S

SV V

K=

+ 1.8

where [ ]max 2 0EV k= and it is the maximum velocity of the enzymatic reaction at

maximum (saturated) concentration of the substrate. 1 2M

1

k kK

k

− += is a Michaelis-

Menten constant (units of mol l−1) and defines a concentration of substrate at half-


39

maximal rate ( max0 2

VV = ) (Figure 1.26). In case of slow product formation, the KM

constant reduces to the form of the dissociation constant KD and can be used as a

rough indication of stability of the intermediate complex [ES] (Equation 1.9). When

the substrate binds weakly with the enzyme, KM will have large values and when the

substrate binds tightly, the KM values will decrease.

1D M

1

kK K

k

−= = , if k2 << k-1 1.9

Figure 1.26 shows a typical plot of the reaction velocity V as a function of the

substrate concentration [S], resembling a hyperbolic curve. In the case when [S] <<

KM, the reaction velocity increases linearly with the substrate concentration and the

Michaelis-Menten equation is reduced to the form

[ ]20

M

ESk

VK

= 1.10

where 2

M

k

Kis known as a specific rate constant and used in biochemistry to compare

the efficiency of different enzymes. Under these conditions, the majority of the

enzyme and substrate are free and reaction is considered as bimolecular and the rate

of the reaction depends on how efficiently the enzyme can bind to substrate at that

concentration. When [S]=KM, the initial reaction velocity is exactly half the

maximum velocity obtained for the enzyme, max0 2

VV = . At high substrate

concentrations, [S] >> KM, all molecules of enzyme are saturated with substrate,

therefore the rate has its maximum value and is independent on the substrate

concentration, maxV V= .


40

V

[S]

V=Vmax

1/2 Vmax

KM

V = Vmax

[S] / KM

Figure 1.26 Relation between reaction velocity V and the concentration of substrate defined by

Equation 1.8.

Values of KM and k2 can be determined from graphical representation of the

reciprocal forms of the Equation 1.8, yielding straight-line plots and these are often

used to obtain KM and k2. One of them is the Lineweaver-Burke equation

M

0 max max

1 1 1[S]

K

V V V= + 1.11

where plotting 1/V as a function of 1/[S] gives a straight line with intercept equals

max

1V

and slope M

max

K

V.

A more common and statistically accurate method to calculate kinetic values of KM

and k2 for enzymatic reactions obeying the Michaelis-Menten model is to use a non-

linear least square fitting routine to directly fit the experimental data to Equation 1.8.

Functionalisation of individual GC electrodes

41

2 Functionalisation of individual GC electrodes

2.1 General

In this project, we are interested in development of new covalently modified

carbon electrodes for application in NADH-dependent biosensors. Methodology

proposed by Kilburn and Bartlett for covalent functionalisation of glassy carbon

(GC) electrodes allows the molecular architecture at the carbon surface to be built up

using electrochemical and solid-state synthesis methods (Scheme 2.1).56,57,136

According to their approach, the initial step of the electrode functionalisation

involves electrochemical oxidation of amines or electrochemical reduction of

diazonium salts at the electrode surface, leading to formation of a monolayer of the

linker at the GC surface. The mono-Boc protected amine group of the linker allows

for subsequent functionalisation of the GC electrodes using solid-phase synthetic

methods. This novel approach towards modification of GC electrodes was

successfully applied for introduction of a number of redox active species, including

well-known NADH mediators, such as dihydroxybenzene derivatives.56

Scheme 2.1 General approach for sequential covalent modification of the GC electrodes developed by

Kilburn and Bartlett;57 a) electrochemical reduction of Boc protected amine diazonium salts or

oxidation of mono-Boc protected primary diamine; b) removal of the Boc protecting group; c) solid-

phase coupling reaction of a redox center at the GC surface.

In this project, we were interested in applying the modification method

presented in Scheme 2.1 for covalent attachment of quinone containing structures,

reported in the literature as effective mediators for NADH oxidation.56,99,109 An

interesting group of such mediators contain 1,10-phenanthroline-5,6-dione

(phendione), a bidentate ligand, coordinated to transition metal ions, which is known

to catalyse NADH oxidation at low overpotential when the metal complexes are

physically adsorbed at the carbon electrode.127,130-134 Despite strong adsorption of the

phendione at a carbon surface, there was no molecular control over the modified


42

carbon surface and consequently a risk of decrease in the catalytic activity of the

adsorbed metal complex due to desorption from the electrode surface. In order to

provide molecular control over the metal complexes with phendione ligand(s) and

increase their stability at the electrode surface, the strategy of covalent attachment of

the phendione containing metal complexes was proposed (Scheme 2.2). According to

this strategy, the mono-Boc-protected amine linker was electrochemically attached at

the surface followed by coupling of a carboxylic acid-functionalised bidentate ligand

using the method optimised previously by Kilburn and Bartlett.56,57,136 The presence

of the bidentate ligand at the GC surface allowed for formation of different metal

complexes with the phendione redox active ligand(s). In this way the metal complex

containing phendione redox active ligand(s) was covalently immobilised at the

electrode surface.

Scheme 2.2 Proposed strategy of covalent immobilisation of metal complexes at the GC electrode

using electrochemical and solid-phase coupling conditions.

In this Section, the synthesis of three different octahedral ruthenium complexes

containing phendione ligand(s) will be described followed by their electrochemical

characterisation in homogenous solution and their covalent attachment to GC

electrodes according to the proposed strategy (Scheme 2.2). In order to study the

effect of geometry and size of the metal complex, immobilisation of tetrahedral

(1,10-phenanthroline-5,6-dione)zinc (II) chloride at the GC electrode was also

proposed according to the strategy depicted in Scheme 2.2.

The proposed general strategy of functionalisation of GC electrodes by metal

complexes allowed for variation between different linkers, bidentate ligands and

metal complexes. Thus, this method was used in further work to design the

combinatorial library of modified electrodes followed by HTP screening of the

library towards NADH oxidation (Section 3).


43

2.2 Synthesis of ruthenium complexes with 1,10-phenanthroline-

5,6-dione chelating ligand.

Octahedral ruthenium metal complexes with different chelating ligands were

synthesised using the well know dichlorotetrakis(dimethyl sulphoxide) ruthenium

(II) precursor 1, prepared according to the literature procedure.137 Reduction of the

hydrated ruthenium (III) chloride in neat DMSO gave octahedral complex 1 with

chloride ligands in cis positions (Scheme 2.3). The structure and geometry of

precursor 1 was difficult to determine by NMR due to the fact that the very labile

DMSO ligands exchange rapidly with deuterated solvent. A geometrical

configuration of the precursor 1 was confirmed by X-ray and clearly shows that the

chloride ligands are in cis configuration and three molecules of DMSO are

coordinated via sulfur and one is coordinated through oxygen. This is in good

agreement with the literature data, where the distortion in the structure of complex 1

was explained by the steric effect of the methyl groups in DMSO, which prevents all

of the DMSO ligands being bonded through sulfur to the ruthenium metal centre.

RuCl

Cl SOMe2

RuCl3 nH2O+SOMe2

OSMe2

SOMe2

dmso

1

a

Scheme 2.3 Preparation of the precursor 1 and its X-ray crystallographic structure (right); a) 5 min

reflux in neat DMSO under nitrogen, acetone, 52 %.

The precursor 1 was used for synthesis of different octahedral ruthenium complexes

with the phendione chelating ligand 2, which was prepared according to the literature

procedure in 60 % yield (lit.138 86 %). Firstly, two labile DMSO ligands in precursor

1 were replaced with one ligand 2 by mixing one equivalent of the precursor 1 with

one equivalent of the ligand 2 in inert solvent. Literature reported reactions between

precursor 1 and different nitrogen and phosphorous ligands were performed in

nonpolar inert solvents, such as chloroform and toluene.139 In this case, ethanol was


44

chosen as the reaction solvent due to low solubility of the ligand 2 in non-polar

solvents. The reaction progress was monitored by 1H NMR at intervals of one hour.

After 14 h the reaction was completed and characteristic protons of the starting

materials 2 were not present in the 1H NMR spectrum. The product was precipitated

as a brown microcrystalline solid. Literature examples of the precursor 1 with

various nitrogen and phosphorus chelating ligands indicated that displacement of

weaker O-bonded DMSO ligand in precursor 1 presumably occurs first, followed by

displacement of a S-bonded DMSO ligand in the trans position.137,140 The precursor 1

obtained as the cis isomer in respect to two chloride ligands led to formation of the

complex 3 also in the cis geometrical configuration, which was confirmed by NMR

analysis. The 1H NMR spectrum showed the presence of six resonance peaks in the

aromatic region, which correspond to six non-equivalent protons of the phendione

ligands in the cis position to chlorine and DMSO ligand and the other site being in

the trans position to the other monodentate ligands. In addition, three singlets

between 3.43 and 2.33 ppm, corresponding to twelve protons of methyl groups,

confirm the presence of DMSO ligands in the complex 3.

Scheme 2.4 Preparation of complex 3; a) precursor 1, ethanol, reflux, 14 h, 60 %.

Literature examples of the metal complexes containing redox-active phendione

ligands were studied for their electrocatalytic activity towards NADH oxidation.

Abruña et al.127 reported preparation and evaluation of the electrochemical and

catalytic properties of complexes of general form [M(phendione)3]2+. Cyclic

voltammetry of these complexes in aqueous solution showed reversible

electrochemical behaviour corresponding to quinone functional groups at middle

peak potential Emp of 0.15 V vs. SSCE.127 A similar result was observed by cycling a

1 mM aqueous solution of the complex 3 (Figure 2.1) in the potential range from

−0.5 to 0.3 V vs. SCE, which shows the reversible redox process of the quinone

functionality at a middle peak potential Emp of −0.18 V vs. SCE. The presence of the


45

peak corresponding to the RuII/RuIII redox couple was not observed in the chosen

potential range. According to studies by Lever141 coordination of monodentate

ligands such as DMSO shifts the potential of the ruthenium RuII/RuIII redox couple to

more positive values of about 1.2 V vs. SHE in aqueous solution. In the case of

carbon electrodes this peak might overlap with the region corresponding to oxidation

of the carbon surface.

Figure 2.1 Cyclic voltammogram of 1 mM solution of complex 3 in 0.1 M phosphate buffer solution

pH 7 at scan rate of 50 mV s−1 and GC electrode geometrical surface area of 0.071 cm2.

The cis configuration of chlorine atoms in complex 3 allowed for isomerically

controlled substitution of the remaining two DMSO ligands in 3 by one molecule of

2,2’-bipyridyl chelating ligand, using reaction conditions proposed by Grätzel et

al..142 Complex 3 was mixed with one equivalent of the 2,2’-bipyridine ligand in

DMF and the reaction mixture was heated under reflux (Scheme 2.5). DMF was

chosen as a reaction solvent in order to increase solubility of the complex 3 and the

reaction temperature in order to apply higher energy required to displace two DMSO

ligands by one more stable 2,2’-bipyridyl chelating ligand.140 Undesired cis-trans

isomerisation during complex formation was prevented by performing the reaction

under reduced light. Pure cis-(2,2’-bipyridyl)(1,10-phenanthroline-5,6-

dione)ruthenium (II) chloride complex 4 was obtained after precipitation from

methanol. The cis geometry of complex 4 was confirmed by NMR analysis. No

resonance peaks corresponding to methyl groups of the DMSO ligands were

observed, which would suggest that two remaining DMSO ligands in the complex 3

-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3-15

-10

-5

0

5

10

i /

µA

E vs. SCE / V


46

were completely replaced by one 2,2’-bipyridyl ligand. The 1H NMR spectrum

indicated that the two chelating ligands are trans, which was indicated as fourteen

different resonance peaks in the aromatic region.

Scheme 2.5 Synthesis of the complex 4; a) 2,2’-bipyridine, DMF, reflux 4 h, reduced light, 60 %.

Complex 4 was electrochemically characterised in aqueous and non-aqueous

solutions (Figure 2.2). The cyclic voltammogram of complex 4 in phosphate buffer

solution indicates the presence of a reversible redox couple at a value of Emp of −0.06

V vs. SCE, which corresponds to the 2e- redox process of quinone in the phendione

ligand.127 A broad, reversible redox peak at potential of 0.35 V vs. SCE suggests the

presence of RuII/RuIII redox couple, which is in good agreement with the theoretical

potential values of 0.3 V vs. SHE calculated for the RuII/RuIII redox center in similar

metal complexes according to Lever.141 Figure 2.2b shows the cyclic voltammogram

of the complex 4 in acetonitrile solution, which clearly indicates the presence of two

reversible redox peaks at −0.05 and −0.7 V vs. Ag/AgCl and corresponds to the one

electron, two step redox process of the phendione ligand. At a relatively positive

potential of 0.6 V vs. Ag/AgCl, one electron redox process of RuII/RuIII was

observed. In addition, at a potential of −1.5 V vs. Ag/AgCl, a redox couple was also

observed, which might correspond to the 2,2’-bipyridyl ligand. Results of the

electrochemical screening of complex 4 in non-aqueous solution are in good

agreement with those reported by Abruña et al. for similar bis and tris ruthenium

complexes.127


47

-0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2

-20

-10

0

10

20

30

i /

µA

E vs. SCE / V

a)

Figure 2.2 Cyclic voltammograms of 1 mM solution of complex 4 at GC electrode (geometrical

electrode area of 0.071 cm2); a) in 0.1 M phosphate buffer solution pH 7 at scan rate of 50 mV s−1; b)

in acetonitrile with 0.1 M TBATFB at scan rate of 50 mV s−1.

Finally, precursor 1 was used for synthesis of the octahedral complex bis(1,10-

phenanthroline-5,6-dione)ruthenium (II) chloride 5. In this case, addition of two

equivalents of the ligand 2 to one equivalent of the precursor 1 in DMF was found to

be an effective method for displacement of four DMSO ligands in 1 with two

phendione chelating ligands in a single reaction step (Scheme 2.6). Complex 5 was

successfully prepared using the same reaction conditions as for formation of the bis

complex 4 and obtained in 33 % yield as the cis isomer with respect to the chlorine

ligands. The geometry of complex 5 was confirmed by NMR analysis. The 1H NMR

spectrum of 5 shows six resonance peaks in the aromatic region, which correspond to

six non-equivalent proton of two mutually adjacent sites of the trans phendione

ligands.

Scheme 2.6 Preparation of complex 5; a) precursor 1, DMF, reflux 4 h, 33 %.

The precipitated complex 5 shows very low solubility in aqueous and non-aqueous

solvents at room temperature, therefore electrochemical characterisation of complex

5 in either aqueous or non-aqueous solution was unsuccessful.

-2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0-30

-20

-10

0

10

20

30

i /

µA

E vs. SCE / V

b)


48

2.3 Attachment of bis-(dimethyl sulphoxide)(1,10-phenanthroline-

5,6-dione) ruthenium (II) chloride complex at the GC surface.

Ruthenium complexes 3, 4 and 5 were applied in the following work for

functionalisation of GC electrodes according to the strategy proposed in Scheme 2.2

and evaluated towards NADH catalytic oxidation. The first part of this work

involved optimisation of solid-phase reaction conditions for attachment of the

complexes 3, 4 and 5 at the GC electrodes. Each GC electrode was polished on dry

silicone-carbide paper (grade 1200) directly prior to modifications, and the polishing

process was carried out in the same manner for all GC electrodes discussed below.

According to the proposed strategy for multistep sequential functionalisation of the

GC electrodes by various metal complexes (Scheme 2.2), the first step of the

modification involved electrochemical attachment of linkers at the blank GC

electrodes by electroxidation of primary amines to create a stable N-C bond between

the electrode surface and the linker.56,57,136 Initially, mono-Boc protected

ethylenediamine (EDA) 6 was chosen as a linker due to relatively high surface

coverage obtained for modifications with the linker 6 reported in the literature.57 The

linker 6 was covalently tethered at the blank GC electrode according to the literature

procedure, resulting in formation of a monolayer of the mono-Boc-EDA at the

electrode surface (Figure 2.3). Linker 6 was oxidised at the GC surface by cycling

the potential in the range 0 to 2.25 vs. SCE. During the initial cycle, a significant

oxidation peak at a potential of 1.8 V vs. Ag/AgCl was observed, corresponding to

formation of an intermediate amine radical in solution and its reaction with carbon

atoms at the GC electrode surface. Decrease of the current with following cycles

indicates that the electrode surface was blocked by the linker 6, resulting in

formation of the full monolayer of 6 at the GC surface.

It is well known that oxidation of the mono-Boc protected diamines at the GC

surface prevents formation of bridge structures, where both amine groups are

attached at the surface.57 In addition, presence of the Boc-protected amine at the GC

surface allows for control over subsequent chemical steps carried out under solid-

phase conditions.


49

0.0 0.5 1.0 1.5 2.0 2.5

0

50

100

150

200

250

i /

µA

E vs. SCE / V

EDA

Figure 2.3 Oxidation of linker 6 at the polished GC electrode with geometrical area of 0.071 cm2;

general reaction scheme and cyclic voltammogram for attachment of mono-Boc-EDA linker 6 (below)

recorded in 10 mL of 15 mM solution of 6 in acetonitrile with 0.1 M TBATFB, from 0 to 2.25 V vs.

Ag/AgCl at scan rate of 50 mV s−1.

The Boc protecting group was cleanly removed from the modified electrode 7

using the literature procedure,57 resulting in free primary amine groups at the

electrode surface. According to the proposed strategy (Scheme 2.2) the free amine

groups allowed for subsequent introduction of a chelating ligand at the GC surface

under solid-phase conditions for amide formation. Initial work involved

functionalisation of the GC electrodes with 2,2’-bipyridyl ligand, known to create

stable metal complexes with a large number of transition metals. In this work, the

2,2’-bipyirdine chelating ligand was functionalised with a carboxylic acid group 9 in

order to covalently immobilise the ligand at the modified electrode 7. Toy et a.l143

reported preparation of the 2,2’-bipyridine chelating ligand 9 using Kröhnke

synthesis of unsymmetrically functionalised 2,2’-bipyridines (Scheme 2.7).144

Scheme 2.7 Synthesis of 2,2-bipyridine-5-carboxylic acid 9 according to the literature procedure

reported by Toy et al.143; a) pyridine, iodine, 90 °C, 6 h, 55 %; b) methacrolein, NH4OAc, formamide,

80 °C, overnight, 90 %; c) potassium permanganate, H2O, HCl, 70 – 90 °C, 7 h, 55 %.

Ligand 9 was successfully obtained in three steps using this literature procedure in

60 % yield (lit.143 80 %) and covalently immobilised at the modified electrode 8


50

under solid-phase peptide coupling conditions using HBTU as a coupling reagent.

The modified electrode 10 was prepared under solid-phase reaction conditions

optimised for covalent attachment of anthraquinone-2-carboxylic acid (AQ) as

reported by Kilburn and Bartlett.56,57,136

Scheme 2.8 Removal of Boc protecting group from electrode 7 followed by solid-phase coupling of

the ligand 9; a) 4.0 M hydrochloric acid in 1,4-dioxane, 1 h, b) HBTU, DIPEA, DMF, 16 h, room

temperature.

The yield of solid-phase coupling of the ligand 9 to the electrode 8 was difficult to

determine directly by electrochemical methods. Cyclic voltammograms of the

electrode 10 performed in aqueous and nonaqueous solvents did not show redox

peaks which could correspond to the 2,2’-bipyridine ligand at the GC surface.

Efficiency of this coupling reaction was determined by performing an experiment,

where the modified electrodes 7 and 10 were both dipped in the same solution of

complex 3 and heated overnight. The resulting modified electrodes were screened by

cyclic voltammetry and showed significant presence of the quinone groups at a

potential of 0 V vs. SCE. In case of the electrode 10, a relatively high number of the

quinone groups were observed in comparison with the electrode with mono-Boc

protected EDA linker 7 (Figure 2.4). This would suggest that 2,2’-bipyridine at the

electrode 10 coordinates to the ruthenium metal ion in complex 3, resulting in

formation of a new ruthenium complex at the electrode surface. For electrode 7,

negligible amount of the complex was present at the electrode surface, which might

be rationalised by physical adsorption of this complex at the carbon surface. This

control experiment confirmed that the 2,2’-bipyridine was successfully attached at

the electrode surface and coordinated to the ruthenium metal ion.


51

-0.4 -0.2 0.0 0.2 0.4 0.6-8

-6

-4

-2

0

2

4

6

8

electrode 7

electrode 10

i /

µA

E vs. SCE / V

Figure 2.4 Cyclic voltammograms obtained after heating of electrodes 7 (red) and 10 (black) in a

solution of the complex 3 in 0.1 M phosphate buffer pH 7 at a scan rate of 50 mV s-1 for electrodes

with geometrical surface area of 0.071 cm2.

The final step of functionalisation of the GC electrodes involved coordination

of the ruthenium complexes 3, 4 and 5 to the bipyridyl ligand present at the electrode

10. Initial experiments involved coordination of complex 3 at modified electrode 10.

Solid-phase coordination of the bipyridyl ligand to ruthenium was carried out by

dipping electrode 10 in a solution of complex 3 in DMF. In this case, DMF was

chosen as the reaction solvent due to high solubility of the complex 3 and was

reported in the literature for formation of different tris and bis ruthenium

complexes.139,140 The solid-phase reaction conditions, including temperature,

concentration of 3 solution and reaction time were optimised as shown in Scheme

2.9.

The effect of different reaction conditions on the yield of attachment of complex 3 at

the GC electrode was determined by calculating the surface coverage Γmed of the

redox active quinone groups present at the surface. Values of Γmed were estimated

from the charge (Q) obtained by the integration of the area under the oxidation and

reduction peak in cyclic voltammograms according to Faraday’s law

Γmed = Q/nFAρ 2.1

where n equals 2 is the number of electrons transferred, F is the Faraday’s constant,

A is the electrode area and ρ is the roughness of the electrode. A roughness factor of

4 was used in all calculations as this was reported previously for GC electrodes

polished in the same manner.55,57


52

Firstly, attachment of complex 3 at the GC surface was performed at three different

concentrations of 3 (0.5, 0.1, 0.01 M). Calculated values of Γmed increased with lower

concentration of complex 3 (Scheme 2.9). The effect of the reaction temperature on

values of Γmed was evaluated by performing the solid-phase coordination reaction at

three different temperatures. It was observed that coordination of the 2,2’-bipyridyl

present at electrode 10 to the ruthenium metal ion in complex 3 at room temperature

was not effective, the cyclic voltammogram of modified electrode 11 prepared at

room temperature did not show peaks corresponding to the quinone groups in 3.

Increasing the temperature of the solid-phase coordination process to 80 or 100 °C

for modified electrodes 12 and 13 induced the coordination of the 2,2’-bipyridine to

the ruthenium metal centre in 3. Reversible redox processes at 0 V vs. SCE were

observed for both and values of Γmed of 0.08 and 0.09 nmol cm−2 were estimated for

12 and 13, respectively. For modified electrodes 14 and 15, where solid-phase

reactions were carried out under reflux in DMF for 4 and 16 h, cyclic

voltammograms did not show the redox peaks expected for the quinone redox

process. The high temperature of the reaction mixture might cause decomposition of

the EDA linker at the surface and prevents coordination of 3 to the surface.

From data on octahedral osmium complexes, coverage of 1 × 10-10 mol cm-2

represents a monolayer of the osmium complex at the electrode surface.145 The value

of Γmed obtained for modified electrode 13 is closest to the theoretical literature value

of a full monolayer of the octahedral complex. For this reason, the reaction

conditions applied during modification of electrode 13 were used in further work

involving formation of the complex A.


53

Electrode

number

Concentration

of complex 3 / mol L−−−−1

Temperature

/ °C Time / h

Surface coverage Гmed

/ nmol cm−−−−2

11 0.1 R.T. 16 No quinones present

12 0.1 80 16 0.081

13 0.1 100 16 0.087

14 0.1 reflux 4 No quinones present

15 0.1 reflux 16 No quinones present

16 0.5 80 16 0.055

17 0.01 100 16 0.11

Scheme 2.9 Optimisation process of the attachment of the complex 3 at the modified electrode 10

under solid-phase reaction conditions.

Modified electrode 17, which gave the highest calculated values of Γmed, was

evaluated towards catalytic oxidation of NADH. The cyclic voltammogram (Figure

2.5) of electrode 17, recorded in presence of 1 mM NADH, showed a catalytic

anodic peak at 0 V vs. SCE. This suggests that the phendione at the surface reduced

the overpotential by approximate 0.4 V relative to oxidation of NADH at the bare

carbon electrode occurring at a potential of 0.4 V vs. SCE.93,146 Presence of NADH

peak was shifted by about 40 mV to more positive potential in comparison to the

potential of the mediator. This might be caused by slow reaction rate between

quinones at the surface and NADH in solution.97,110


54

-0.25 -0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20

-4

-2

0

2

4

6

8

electrode 13 without NADH

electrode 13 with 1 mM NADH

i /

µA

E vs. SCE / V

Figure 2.5 Cyclic voltammograms of modified electrode 17 in the presence (red) and absence (black)

of 1 mM NADH in 0.1 M phosphate buffer solution pH 7 at a scan rate of 50 mV s-1 (geometrical

electrode area of 0.071 cm2).

Based on the successful preparation of the initial modified electrode 17, the

same reaction conditions were applied for covalent attachment of the complex 3 at

the GC surface through six different linkers in order to investigate the effect of the

structure and length of the linker on the catalytic activity of complex A towards

NADH oxidation.

The choice of linker was based on work reported by Bartlett et al.57

describing

functionalisation of GC electrodes by AQ redox probe covalently attached at the

surface through linkers with various aliphatic chain lengths and structures.

Electrochemical characterisation of AQ modified electrodes clearly showed that

length and structure of the linker have an impact on surface coverage of the AQ at

GC surface and electron transfer kinetics between the AQ redox center and electrode

surface.57 To our best knowledge, the effect of different linkers on NADH

electrocatalytic activity for the electrodes modified according to the methodology

developed by Bartlett and Kilburn was not reported.

Therefore, covalent attachment of complex A through linkers with different length

and structures was proposed in order to evaluate effect of linker on electrochemical

and catalytic properties of complex A at the surface.

Based on the successful preparation of the electrode 17 modified with EDA linker,

aliphatic diamine linkers with increased chain lengths were chosen, including mono-

N-Boc-butanediamine (BDA) 18 and mono-Boc-1,6-hexanediamine (HDA) 19.

Different lengths of the linker might affect accessibility of complex A during

electrocatalytic oxidation of the bulky molecule of NADH in solution. In order to


55

investigate the structure of diamine linkers, hydrophilic mono-Boc-2,2'-

(ethylenedioxy)diethylamine chain (EDDA) 20 and aromatic mono-Boc-p-

xylenediamine (XDA) 21 were selected and applied for covalent attachment of the

complex 3 according to the strategy proposed in Scheme 2.2.

Scheme 2.10 Covalent attachment of linkers 6 and 18-21 at the GC electrodes; a) applied potential

from 0 to 2.25 V vs. Ag/AgCl in 15 mM solution of the linker in acetonitrile with 0.1 M TBATFB at

scan rate of 50 mV s-1.

Mono-N-Boc protected diamines were electrochemically oxidised at the GC

electrode by applying potential range from 0 to 2.25 V vs. Ag/AgCl. Figure 2.6

shows an example of a cyclic voltammogram recorded for oxidation of linker 20 at

blank GC electrode. Similarly as for attachment of the linker 6, an oxidation peak at

1.8 V vs. Ag/AgCl was observed, corresponding to formation of the amine radical,

which reacts with sp2 carbon atoms at GC and leads to a stable carbon-nitrogen bond

at the surface. Decrease of the oxidation current observed in the subsequent scans

suggests blocking of the surface with the linker 20, which is consistent with the result

obtained for modification of the electrode with 6 (Figure 2.3), where a full

monolayer of the EDA was obtained after three scans. The same cyclic


56

voltammograms were obtained during oxidation of diamine linkers 18-21 and full

monolayer of the linker attached at GC surface was obtained after 3-5 cycles.

0.0 0.5 1.0 1.5 2.0 2.5

0

20

40

60

80

i /

µA

E vs. SCE / V

EDDA

Figure 2.6 Cyclic voltammogram recorded in 15 mM solution of linker 20 in acetonitrile with 0.1 M

TBATFB at scan rate of 50 mV s-1 at GC surface with geometrical area of 0.071 cm2.

The versatile methodology for functionalisation of the GC electrodes

developed by Bartlett and Kilburn also included covalent attachment of linker using

electrochemical reduction of a diazonium salt, which is well known in the literature

as an effective method for formation of stable C-C bond between organic molecules

and the carbon surface.64,65,73,74 Electrochemical reduction of 4-(N-Boc-

aminomethyl)benzene diazonium tetrabluoroborate salt allowed for covalent

immobilisation of the AQ according to the methodology developed by Bartlett and

Kilburn (Scheme 2.1).55 In this project, diazonium salt linker 27 was prepared

according to literature procedures147,148 in two steps synthesis (Scheme 2.11).

Scheme 2.11 Two steps synthesis of diazonium salt linker 27 perpared according to the literature

procedures147,148; a) (Boc)2O, DCM, overnight, room temperature, 46 %; b) NaNO2, 40% solution of

HBF4, 1 h, 45 %.

Linker 27 was electrochemically reduced at the GC electrode by applying the

potential from 0.6 to −1 V vs. Ag/AgCl in non-aqueous solvent. Scheme 2.12 shows

example of the cyclic voltammogram recorded for electrochemical reduction of the

linker 27 where the broad, irreversible reduction peak occurring at a potential of −0.5

V corresponds to formation of the aromatic radical and formation of stable carbon-


57

carbon bond with sp2 atoms at the GC. Subsequent cycling in the same solution

showed disappearance of the cathodic wave and this corresponds to fully blocked

surface by the organic layer. It is believed that presence of the bulky Boc-protecting

group in the C-4 position of the linker 21 prevents formation of multilayers of the

aromatic rings at the electrode surface.65,74,75

-1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8

-10

-8

-6

-4

-2

0

2

4i /

µA

E vs. SCE / V

BA

Scheme 2.12 Attachment of linker 27 at the GC electrode; general scheme and cyclic voltammogram

of the electrochemical reduction of 15 mM solution of 27 in acetonitrile with 0.1 M TBATFB at scan

rate of 50 mV s-1 at GC electrode with geometrical electrode area of 0.071 cm2.

The presence of the protected amine functional groups at the modified electrodes 21-

26 allowed for following functionalisation of these electrodes under solid-phase

conditions according to the strategy proposed in Scheme 2.2 in order to covalently

immobilise complex 3 at the GC electrodes. Three subsequent solid-phase synthetic

steps, involving removal of the Boc protecting group, coupling of ligand 9 and

formation of complex A at the electrode surface, were carried out under the same

reaction conditions as for preparation of electrode 17 (Scheme 2.13). Each of the

electrodes 22-26 and 28 were functionalised individually in separate reaction vessels

to give electrodes 17 and 35-39, modified with six different linkers, ligand 9 and

complex A.


58

Scheme 2.13 Solid-phase modifications of the electrodes 22-26 and 28; a) 4.0 M hydrochloric acid in

1,4-dioxane, b) ligand 9, DMF, HBTU, DIEA, room temperature, 16 h, c) complex 3, DMF, 100 °C,

16 h.

The modified electrodes 17 and 35-39 were individually characterised using

cyclic voltammetry in phosphate buffer solution pH 7. Firstly, each modified

electrode was cycled over potential range −0.2 to 0.15 V vs. SCE (Figure 2.7). A two

electron, reversible redox process was observed for all electrodes modified with the

different linkers. Minor changes were observed for the redox midpeak potentials

(Emp) between electrodes 17 and 35-39 with an average values of −50 mV vs. SCE.

Thus, the type of linker does not have a significant effect on thermodynamics for the

redox process of complex A at the electrode surface. Similarly, only minor changes

were observed in peak potential separation (∆Ep) between modifications 17 and 35-

39 with an average value of 40 mV vs. SCE between different linkers at scan rate of

50 mV s-1. This would suggest that the type of linker does not affect the kinetics of

electron transfer between the GC surface and complex A.


59

-0.2 -0.1 0.0 0.1 0.2-4

-2

0

2

4

6

i /

µA

E vs. SCE / V

EDA

-0.2 -0.1 0.0 0.1 0.2

-4

-2

0

2

4

6BDA

i /

µA

E vs. SCE / V

-0.2 -0.1 0.0 0.1 0.2

-4

-2

0

2

4

6

HDA

i /

µA

E vs. SCE / V

-0.2 -0.1 0.0 0.1 0.2

-4

-2

0

2

4

6

8

XDAi /

µA

E vs. SCE / V

-0.3 -0.2 -0.1 0.0 0.1 0.2-4

-2

0

2

4

6EDDA

i /

µA

E vs. SCE / V

-0.2 -0.1 0.0 0.1 0.2-4

-2

0

2

4

6BA

i /

µA

E vs. SCE / V

Figure 2.7 Examples of cyclic voltammograms of the modified electrodes 17 and 35-39 in the

presence (red) and absence (black) of 1 mM NADH in 0.1 M phosphate buffer solution pH 7 at scan

rate of 50 mV s-1 (geometrical electrode area of 0.071 cm2).

The efficiency of formation of complex A attached through different types of

linkers was compared by calculations of surface coverages, Γmed, of the phendione

using Equation 2.1. Figure 2.8 shows a summary of the values of Γmed calculated for

electrodes 17 and 35-39. For aliphatic diamine linkers, values of Γmed tend to

decrease with increasing length of the aliphatic chain, with the highest coverage

obtained for EDA and BDA linkers. Relatively low values of Γmed were found for

electrodes modified with long HDA and EDDA linkers. This might be rationalised

by the fact that the long aliphatic chain in the diamine linker might cause a

conformational disorder within the chains and hinder the access of the free amine

group to the carbon surface and prevent the amine from being covalently attached to

the GC surface.149 This might result in lower coverage in the subsequent steps of

solid-phase coupling of 9 and coordination of complex 3. In addition, relatively low

values of Γmed were obtained for the aromatic diamine linkers XDA and BA. The

lowest value of coverage was observed for the complex A attached at the surface

through the benzylamine linker (BA). The rigid structure of the aromatic ring directly

attached at the GC might limit access of the 2,2-bipyridine to bulky complex A,

resulting in low coverage of complex A.


60

EDA EDDA XDA BDA HDA BA0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08 complex A

Γm

ed /

10

-9 m

ol cm

-2

linker

Figure 2.8 Values of Γmed calculated for modified electrodes 17 and 35-39; Each bar represents mean

values of Γmed calculated according to Equation 2.1. Each bar represents mean value of Γmed and

standard error of the mean of three replicate modified electrodes.

Modified electrodes 17 and 35-39 were evaluated one by one towards NADH

catalytic activity by cycling the electrodes in 1 mM NADH solution at pH 7. In the

presence of NADH, all of electrodes 17 and 35-39 show a catalytic anodic peak at 0

V vs. SCE, which is consistent with result obtained for initial experiments at

electrode 17. Values of icat were measured as a difference between the current with

NADH and the current without NADH at potential of 0.15 V vs. SCE, where all of

the mediator is in its oxidised form and allowed for measurement of the catalytic

current generated only by NADH.

In general, significant values of icat were recorded for modifications with

diamine linkers and they are comparable with values of icat found for modification

with XDA and EDDA. The lowest value of icat was recorded for the modification

with BA, which might limit access of the phendione mediators at the electrode

surface to the bulky NADH molecule in the solution. The values of icat recorded for

the electrode 17 and 35-39 strongly depends on surface coverage of the phendione.

In order to calculate efficiency between the modifications 17 and 35-39, normalised

catalytic currents inorm were calculated according to equation

catnorm

med[NADH] Γ

ii =

2.2

where [NADH] is the concentration of the NADH in solution, Γmed is the surface

coverage and icat is the catalytic current. As significant value of inorm was obtained for

the electrodes 36 and 37 with long HDA and EDDA linkers, which might enhance


61

access of the phendione to the folded, bulky molecule of the NADH in solution. In

addition, the hydrophilic character of the EDDA linker might enable better

interactions of the anionic NADH molecule and surface of the modified electrode 37.

Significant catalytic efficiency was also observed for the electrode with XDA linker,

which would indicate that the aromatic structure of this linker increases catalytic

efficiency of electrode 38 during NADH oxidation.

EDA BDA HDA XDA EDDA BA0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0 complex A

i cat /

µA

linker

a)

Figure 2.9 a) Catalytic current icat for modified electrodes 17 and 35-39 recorded in 1 mM NADH

solution in 0.1 M phosphate buffer pH 7 at scan rate of 50 mV s-1; b) Normalised catalytic currents

inorm for the modified electrodes 17 and 35-39 calculated according to Equation 2.2; Each bar

represents mean values of icat (a) or inorm (b) calculated for a single modified electrode.

In summary, complex 3 was successfully attached at GC through six different

linkers according to the proposed strategy, involving electrochemical and solid-phase

methodology. All of the modified electrodes showed the presence of the quinone

groups, which confirmed formation of the complex A at the electrode surface. A

series of the electrodes modified with the complex A attached at the surface through

six different linkers showed that length and structure of the linker have an effect on

the surface coverage of the complex A and its catalytic activity towards NADH

oxidation.

EDA BDA HDA XDA EDDA BA0

20

40

60

80

100

120

complex A

i no

rm /

10

6 A

cm

5/m

ol2

linker

b)


62

2.4 Attachment of (2,2’-bipyridine)(1,10-phenanthroline-5,6-dione)

ruthenium (II) chloride complex at the GC surface.

In order to investigate the effect of different ligands coordinated to the

ruthenium metal centre on the NADH catalytic activity of the phendione ligand,

complex 4 was covalently attached at the electrode surface according to the strategy

proposed in Scheme 2.2. This strategy involved three steps electrochemical and

solid-phase synthesis and was successfully applied for modification of the GC

electrodes with complex 3 (Section 2.1). Firstly, electrode 10 bearing the EDA linker

and ligand 9 was prepared under reaction conditions reported in Section 2.3. The

presence of 2,2’-bipyridine ligand at the electrode surface enabled formation of the

stable, tris ruthenium complex B at the GC by coordination of the 2,2’-bipyridine to

the ruthenium metal centre in complex 4 (Scheme 2.14).

Reaction conditions for the final step involving solid-phase formation of complex B

on the surface were optimised. Literature reported formation of tris ruthenium

complexes, where two chloride ligands are displaced by one chelating ligand, were

performed under reflux in order to force exchange of the chloride ligands by one

molecule of 2,2’-bipyridine.142 Solid-phase coordination at electrode 10 was carried

out in 0.01 M solution of 4, based on results obtained for optimisation of reaction

conditions for attachment of 3 (Section 2.3). The formation of complex B at

electrode 10 was initially performed by dissolving complex 4 in a mixture of

methanol and water (1 : 1) and heating under reflux overnight.142 Electrochemical

characterisation of electrode 39 indicated that complex B was formed at the GC

surface with relatively low coverage of 0.5 ×10-10 mol cm-2 in comparison with the

theoretical coverage of 1 × 10-10 mol cm-2 reported for related octahedral osmium

complexes.145 Formation of complex B in dry DMF solvent for the same reaction

time and maintaining reaction temperature at 100 °C gave a coverage of 0.072 mol

cm-2 for modified electrode 40. This might suggests that presence of water in the

reaction decreases the yield for the attachment of complex B at electrode 10.


63

Electrode

number Solvent Temperature/ °C Surface coverage Гmed / nmol cm

−−−−2

40 methanol/water (1 :1) Reflux 0.055

41 dry DMF 100 0.072

Scheme 2.14 Optimisation process of the synthesis of complex B at modified electrode 10.

Modified electrode 41 was evaluated for electrocatalytic activity by screening in 1

mM aqueous solution of NADH. The cyclic voltammetry of electrode 41 in the

presence of NADH clearly shows a catalytic anodic process at 0 V vs. SCE, which is

in good agreement with the potential obtained for complex A and suggests

electrocatalysis occurring during NADH oxidation (Figure 2.10).

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

electrode 41

electrode 41 with 1 mM NADH

i /

µA

E vs. SCE / V

Figure 2.10 Example of cyclic voltammograms of modified electrode 41 recorded in the presence

(red) and absence (black) of 1 mM NADH solution in 0.1 M phosphate buffer pH 7 at a scan rate of

50 mV s-1 and geometrical electrode area of 0.071 cm2.

In order to investigate the effect of different linkers at the surface coverage and

NADH catalytic activity of complex B, complex 4 was coordinated at modified

electrodes 29-34 under solid-phase conditions as optimised for modified electrode

41. Choice of the linker and functionalisation of the GC electrodes with these linkers

was described in detail in Section 2.3. The final step of coordination of complex 4 at

electrodes 29-34 was carried out under reaction conditions as optimised for electrode

0.01 M solution of complex 4,

NH

HN10

complex B

N N

O

RuN

NO

O

N

N

2Cl-+2

100 oC, 16 h

various solvents (table)


64

40 (Scheme 2.15). Each electrode was prepared in individual reaction vessels and

characterised one by one using cyclic voltammetry.

Scheme 2.15 Attachment of complex 4 at electrodes 29-34 functionalised by six different linkers;

Each modified electrode 41-46 was prepared individually under the same solid-phase reaction

conditions a) 0.01 M of complex 4 in dry DMF, 100 °C, 16 h.

Firstly, electrodes 41-46 were electrochemically screened in phosphate buffer over

potential range −0.2 to 0.2 V vs. SCE and reversible, two electrons redox process

was observed for all of the modified electrodes (Figure 2.11), which confirms

presence of immobilised phendione containing complex B at the electrode surface.

Minor changes in values of Emp were observed for modifications with complex B

attached at the surface through six different linkers with calculated average value of

Emp −65 mV vs. SCE. This result is consistent with Emp calculated for series of

modified electrodes with complex A, where minor differences in the middle peak

potential between modifications 41-46 would suggest that changes in structure and

length of the linker does not affect thermodynamics of electron transfer between

complex B containing phendione and electrode surface. In addition, similar values of

Emp obtained between series of modified electrodes A and B would indicate that

presence of different ligands, chloride in complex A and 2,2’-bipyridyl in complex

B, does not appear to have a significant effect on the thermodynamics of redox

process of the phendione redox centre.

In order to compare effect of different linkers on kinetics of electron transfer between

electrode surface and complex B, potential peaks separation ∆Ep were calculated for

electrodes 41-46. Similarly as for series of electrodes with complex A, minor

differences in ∆Ep were observed between 41-46. This clearly suggests, that length

and structure of linker does not affect the kinetics of electron transfer for complex B


65

attached at the surface through six different linkers. In comparison with series of

modified electrodes with complex A, the presence of 2,2’-bipyridine in complex B

decreases the kinetics of electron transfer with an average value of ∆Ep of 50 mV for

electrodes 41-46.

-0.2 -0.1 0.0 0.1 0.2

-6

-4

-2

0

2

4

6

8i /

µA

E vs. SCE / V

EDA

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

BDA

i /

µA

E vs. SCE / V

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

HDA

i /

µA

E vs. SCE / V

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

XDA

i /

µA

E vs. SCE / V

-0.3 -0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

EDDA

i /

µA

E vs. SCE / V

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

BA

i /

µA

E vs. SCE / V

Figure 2.11 Examples of cyclic voltammograms recorded for the electrodes 40-45 modified with

complex B attached at the GC surface through six different linkers in presence (red) and absence

(black) of 1 mM NADH solution in 0.1 M phosphate buffer pH 7 at a scan rate of 50 mV s-1 and

geometrical electrode area of 0.071 cm2.

The efficiency of the formation of complex B attached at the surface through

different linkers was compared by calculation of the surface coverage Γmed according

to Equation 2.1. In general, values of Γmed calculated for electrodes 41-46 are

relatively low in comparison with the coverage obtained for modifications with

complex A. This might be rationalised by the displacement of two chloride ligands

by bulky 2,2’-bipyridine ligand in complex B, causing steric hindrace and decreasing

the amount of complex 4 coordinated to 2,2’-bipyridine at electrode 10. The effect of

the type of linkage on values of Γmed for electrodes modified with complex B was in

good agreement with coverages calculated for the series of modified electrodes A

(Section 2.3). For short aliphatic linkers EDA and BDA, values of Γmed are relatively

high in comparison with the coverages obtained for longer HDA and EDDA linkers


66

(Figure 2.12). Relatively low values of Γmed were observed for modifications with the

aromatic linkers XDA and BA.


0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040 complex B

Γm

ed /

10

-9 m

ol cm

-2

linker

Figure 2.12 Values of surface coverage Γmed of complex B calculated for modified electrodes 41-46

based on experimental data obtained from the cyclic voltammograms depicted in Figure 2.11. Each

bar represents mean values of Γmed calculated according to Equation 2.1. Each bar represents value of

Γmed obtained for a single modified electrode.

Each of the modified electrodes 41-46 was evaluated towards electrocatalytic activity

for NADH oxidation by cycling the electrodes one by one in 1 mM NADH solution.

For all electrodes 41-46 a catalytic anodic peak was observed at 0 V vs. SCE. Values

of catalytic currents icat measured at 0.15 V vs. SCE were compared between

electrodes 41-46 (Figure 2.13a). Significant values of icat were observed for the short

aliphatic EDA and BDA linkers and decrease with increasing length of the linkers as

was observed for modifications 43 and 44. There was minor difference in catalytic

currents recorded for the two aromatic linkers XDA and BA.

The catalytic efficiency of electrodes 41-46 was evaluated by calculation of the

normalised catalytic currents inorm as shown for the series of modified electrodes A,

which was calculated according to Equation 2.2. Significant catalytic efficiency

was obtained for the electrode with complex B attached at the surface through the

long EDDA linker, which would suggest that the long polyethoxy linker provides

better accessibility of the phendione ligands at the surface to bulky molecules of the

NADH in solution (Figure 2.13b). In contrast, relatively low values of inorm were

calculated for electrode 46, modified with the BA linker, where the rigid structure of

the linker might limit access of the quinone redox group to active site of the folded,

bulky molecule of NADH in solution. Despite relatively low coverages obtained for


67

modified electrodes 41-46, values of inorm obtained for complex B are significantly

higher in comparison to values of inorm calculated for the series of modified

electrodes with complex A. This suggests that the presence of the chelating ligand in

the complex B increases the catalytic activity of the modified electrodes towards

NADH oxidation.


0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5 complex B

i ca

t / µ

A

linker

a)

Figure 2.13 a) Catalytic currents icat recorded for modified electrodes 41-46 in the presence of 1 mM

NADH in 0.1 M phosphate buffer solution pH 7 at a scan rate of 50 mV s-1; b) Normalised catalytic

currents inorm calculated for electrodes 41-46 according to Equation 2.2. Each bar represents values

of icat (a) or inorm (b) obtained for single modified electrode.


50

100

150

200

250

300 complex B

i no

rm /

10

6 A

cm

5m

ol-2

linker

b)


68


69

2.5 Attachment of bis-(1,10-phenanthroline-5,6-dione)2 ruthenium

(II) chloride complex at GC surface.

Based on the successful preparation of the series of electrodes modified with

complexes A and B, complex 5 bearing two phendione redox active ligands was also

immobilised at the GC surface according to the strategy proposed in Scheme 2.2.

Firstly, modified electrode 10 was prepared by electrochemical attachment of EDA

linker following by coupling of ligand 9 under solid-phase peptide coupling

conditions (Scheme 2.8).

During the final step, the two chloride ligands in complex 5 were replaced by the

2,2’-bipyridine at electrode 10 and the novel tris ruthenium complex C was formed

at the GC. Reaction conditions optimised for preparation of the similar tris complex

B were also applied for formation of complex C (Scheme 2.15), bearing two

phendione redox active ligands. The cyclic voltammogram in Figure 2.14 clearly

shows the presence of a reversible redox process at Emp of 57 mV vs. SCE,

corresponding to oxidation/reduction of the phendione redox centre. The coverage,

Γmed, of complex C at electrode 47 was calculated using Equation 2.1. The calculated

value of Γmed of 0.059 nmol cm-2 (for n = 4) for electrode 46 is in good agreement

with the coverage obtained for electrode 41, modified with EDA linker and tris

complex B. The cyclic voltammetry of electrode 47 in 1 mM NADH solution shows

significant catalytic current at 0 V vs. SCE. This suggests that the phendione in

complex C acts as an effective mediator towards NADH oxidation and decreases the

overpotential of the oxidation of NADH by about 0.4 V, which is in good agreement

with results obtained for complexes A and B. The catalytic current, icat, recorded at

0.15 V vs. SCE is significantly higher than the current obtained for electrode 40

modified with complex B. This would suggest that the increased number of redox

active phendione ligands enhances the catalytic activity towards NADH oxidation.


70

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

10

12

i /

µA

E vs. SCE /V

electrode 47

electrode 47 in 1 mM NADH

Figure 2.14 Attachment of complex 5 at electrode 10; a) 0.01 M of complex 5 in dry DMF, 100 °C,

16 h; Cyclic voltammograms of electrode 47 recorded in the presence (red) and absence (black) of 1

mM NADH solution in 0.1 M phosphate buffer pH 7 at a scan rate of 50 mV s-1 and geometrical

electrode area of 0.071 cm2.

Successful formation of complex C at GC was followed by preparation of a

series of electrodes modified with six different linkers as described for the series of

electrodes with complexes A and B (Scheme 2.16). Each of modified electrodes 47-

52 was prepared in an individual reaction vessel under the same solid-phase reaction

conditions and characterised, one by one, using cyclic voltammetry.


71

Scheme 2.16 Attachment of complex 5 at electrodes 29-34; Each modified electrode 47-52 was

prepared individually under the same solid-phase conditions a) 0.01 M of complex 5 in dry DMF,

100 °C, 16 h.

Figure 2.15 shows cyclic voltammograms for electrodes 47-52 cycled over the

potential range −0.2 to 0.2 V vs. SCE. The quinone redox peak was observed at an

average value of Emp of 50 mV vs. SCE for all of the electrodes. Similarly as for

complexes A and B, the type of linkage does not affect the values of Emp for the

phendione coordinated to the ruthenium metal centre. An average value of peak

separation ∆Ep of 50 mV measured at a scan rate of 50 mV s-1 was obtained for

electrodes modified with the different linkers. This result is in good agreement with

the peak separations obtained for the series of complexes A and B, which would

suggest that the additional phendione ligand in complex C does not have an impact

on the kinetics of electron transfer between the electrode surface and complex C.


72

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

10

12

14

i /

µA

E vs. SCE / V

EDA

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

10

12

14

BDA

i /

µA

E vs. SCE / V

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

10

12

14

HDA

i /

µA

E vs. SCE / V

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

10

12

14

XDA

i /

µA

E vs. SCE / V

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

10

12

14

EDDA

i /

µA

E vs. SCE / V

-0.2 -0.1 0.0 0.1 0.2-6

-4

-2

0

2

4

6

8

10

12

14

BA

i /

µA

E vs. SCE / V

Figure 2.15 Cyclic voltammograms recorded for electrodes 47-52 modified with complex C attached

at GC surface through six different linkers in the presence (red) and absence (black) of 1 mM NADH

solution in 0.1 M phosphate buffer pH 7 at a scan rate of 50 mV s-1 and geometrical electrode area of

0.071 cm2.

The cyclic voltammograms obtained for electrodes 47-52 in phosphate buffer

solution allowed for evaluation of the coverage of complex C and evaluation of the

effect of the type of linkage on values of Γmed. Figure 2.16 presents values of Γmed for

complex C attached at GC through different linkers and clearly shows that significant

coverage was obtained for the short aliphatic linkers EDA and BDA. Decrease of

Γmed was observed for linkers with longer chains EDDA and HDA, which is

consistent with values of Γmed obtained for electrodes modified with complexes A

and B. Relatively low coverage was obtained for modification 52, where the rigid

structure of the BA linker might decrease number of 2,2’-bipyridine ligands

accessible to coordinate to the rather bulky complex 5.


73


0.01

0.02

0.03

0.04

0.05

0.06

complex C

Γm

ed /

10

-9 m

ol cm

-2

linker

Figure 2.16 Values of surface coverage Γmed of complex C calculated for modified electrodes 47-52

based on experimental data obtained from cyclic voltammograms presented in Figure 2.15. Each bar

represents mean values of Γmed calculated according to Equation 2.1. Each bar represents value of Γmed

obtained for a single modified electrode.

Electrodes 47-52 were evaluated towards NADH catalytic activity by cycling each of

the modified electrodes in 1 mM NADH solution (Figure 2.15). Figure 2.17a

presents a summary of the catalytic currents, icat, recorded at 0.15 V vs. SCE for

modifications with six different linkers. Significant values of icat were observed for

electrodes bearing the aliphatic EDA linker and the aromatic linker XDA. However,

relatively low value of icat was found for modification with the BDA linker, which

was not observed for complexes A and B modified with this linker. Low value of icat

was observed for modification 52, where the rigid BA linker might decrease access

of the two phendione ligands in complex C to folded, bulky molecule of NADH in

solution.

In order to compare the catalytic efficiency between electrodes 47-52, normalised

catalytic currents, inorm, were calculated using Equation 2.2. High values of inorm were

obtained for linkers with aliphatic chains such as EDA or EDDA, which suggests

that more flexible linkers would increase access of complex C to the bulky NADH

molecules in solution (Figure 2.17a). However, calculated inorm for modification with

the BDA linker gave relatively low values of inorm in comparison to other linkers.

Similarly as shown for complexes A and B, the rigid BA linker decreases the

electrocatalytic efficiency of complex C at the surface due to steric hindrance of the

bulky complex C, resulting in limited access of the phendione ligands to NADH

molecules in a solution.


74


20

40

60

80

100

120

140

160

complex C

i no

rm /

10

6 A

cm

5m

ol-2

linker

b)



currents inorm calculated for electrodes 47-52 according to Equation 2.2. Each bar represents values

of icat (a) or inorm (b) obtained for single modified electrode.


1

2

3

4

5

6

7

8 complex C

i cat /

µA

linker

a)


75

2.6 Covalent modification of GC electrodes by (1,10-

phenanthroline-5,6-dione)zinc (II) chloride complex.

Successful attachment of ruthenium complexes 3, 4 and 5 and their catalytic

activity towards NADH oxidation at the functionalised GC electrode showed that the

design of carbon electrodes proposed in Scheme 2.2 can be applied for introduction

of different ruthenium metal complexes containing the phendione redox ligand. The

presence of three different ruthenium complexes A, B and C attached at the surface

through six different linkers allowed for studies of different octahedral metal

complexes and different linkers on the catalytic activity of the phendione ligand(s)

towards NADH oxidation.

In order to study the effect of the geometry of complex and different transition metal

on surface coverage and NADH catalytic activity, synthesis of tetrahedral zinc

complex with the phendione ligand at the electrode was proposed using the strategy

shown in Scheme 2.2. Firstly, the tetrahedral (1,10-phenanthroline-5,6-dione)zinc

(II) chloride complex 53 was prepared according to literature procedure reported by

Boghaei et al. and obtained in 60 % yield (lit.150 85 %). The structure of 53 was

confirmed by NMR analysis and compared with literature data. 1H NMR spectrum

showed three resonance peaks in the aromatic region, which corresponds to three

equivalent protons in the phendione. NMR data obtained for 53 are in good

agreement with experimental data reported in the literature.

Complex 53 was covalently tethered at electrode 10 by replacing two chlorine

ligands by one bipyridyl at electrode 10 to form the new, tetrahedral zinc comp00lex

D (Figure 2.18). Initially, the literature procedure for synthesis of (2,2’-

bipyridine)(1,10-phenanthroline-5,6-dione)zinc (II) hexafluorophosphate complex

was applied for formation of complex D at the surface.150 According to this

procedure, silver nitrate salt was added to solution of complex 53 in order to

exchange two chlorine ligands for nitrate groups to increase solubility of 53 in non-

aqueous solvents. Electrode 10 was dipped in a solution of complex 54, where the

labile nitrate ligands were exchanged by more stable 2,2’-bipyridine to give electrode

55. Cyclic voltammetry of the modified electrode 53 shows quinone reversible peaks

at Emp of −70 mV vs. SCE and highly irreversible peaks at a potential of 0.4 V vs.

SCE, which decreased after subsequent cycling in the same solution. The irreversible


76

peak might correspond to trace amounts of insoluble silver chloride in solution of 54,

formed as a byproduct of the ion exchange. Despite filtration of the reaction mixture

before dipping electrode 10, a trace amount of silver chloride might be present in

solution of 54 and adsorbed at electrode 10 and might affect catalytic activity of the

phendione towards NADH oxidation.

-0.4 -0.2 0.0 0.2 0.4 0.6-40

-30

-20

-10

0

10

20

30

40

50

i /

µA

E vs. SCE / V

Figure 2.18 Preparation of electrode 55 under solid-phase reaction conditions based on the literature

procedure; a) silver nitrate, acetonitrile, complex 53, 16 h, 50 °C b) 0.01M complex 54 in acetonitrile,

50 °C, 16 h; Cyclic voltammogram of electrode 55 in phosphate buffer solution pH 7 at a scan rate of

50 mV s-1 and geometrical electrode area of 0.071 cm2.

In order to eliminate the irreversible peak, possibly caused by adsorption of silver

chloride, conditions for solid-phase formation of complex D were optimised by

performing a series of individual solid-phase reactions at electrode 10 with complex

5 at various temperature and reaction times (Figure 2.19). Due to the low solubility

of complex 5 in acetonitrile, dry DMF was chosen as the reaction solvent. Formation

of complex D at the electrode was carried out in 0.01 M solution of 5, which is the

optimum solution concentration as reported for attachment of ruthenium complexes

at GC surface (Section 2.3).


77

In order to optimise the reaction time, individual reaction vessels were prepared

containing two electrodes 10 dipped in solution of 5 and heated at 50 °C varying

reaction times between 30 minutes to 16 hours. Each of the modified electrodes was

characterised by cyclic voltammetry and the surface coverage, Γmed, was calculated

using Equation 2.1. It was observed that values of Γmed increased with reaction time

and significant coverage of 0.18 nmol cm-2 was observed for two electrodes with

reaction time of 16 h. An approximate theoretical value for monolayer coverage of

0.2 nmol cm-2 for complex D was calculated based upon the geometrical area of the

electrode and the size of the zinc complex D.

In order to optimise the reaction temperature for formation of complex D at the

electrode surface, reactions for attachment of complex 52 at electrode 10 were

carried out at three different temperatures. Electrodes modified at various

temperatures were also characterised by cyclic voltammetry and values of Γmed for

each electrode were calculated. It was observed that relatively high coverage close to

the theoretical value for a monolayer of complex D was obtained for electrodes

modified at 50 and 80 °C in comparison to modification carried out at room

temperature (Figure 2.19b). A temperature of 50 °C was chosen for following

experiments due to the fact that the stability of the GC modified electrodes decreases

with temperature as was observed during optimisation of reaction conditions for

attachment of ruthenium complexes (Section 2.3).


78

20 30 40 50 60 70 800.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20 complex D

Γm

ed

/ 1

0-9 m

ol cm

-2

temperatute / oC

b)

Figure 2.19 Optimisation process of formation of complex D at electrode 10 in 0.01 M solution of 53

in DMF a various reaction time and reaction temperatures; a) Values of Γmed for complex D created at

reaction times varying from 30 minutes to 16 h; b) Values of Γmed for complex D created at different

reaction temperatures for 16 h.

Optimised reaction conditions were applied for attachment of complex 52 at

electrodes modified with six different linkers (Scheme 2.17) as described for

ruthenium complexes A, B and C. Each of electrodes 56-61 was prepared

individually and characterised one by one using cyclic voltammetry.

30min 1h 2h 4h 6h 8h 12h 16 h0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20 complex D

Γm

ed /

10

-9 m

ol cm

-2

reaction time / h

a)


79

Scheme 2.17 Attachment of complex 53 at electrodes 29-34, functionalised by six different linkers;

Each modified electrode 56-61 was prepared individually under the same solid-phase reaction

conditions a) 0.01 M of complex 51 in dry DMF, 50 °C, 16 h.

Cyclic voltammograms of electrodes 56-61 show the presence of reversible

peaks in aqueous solution at an average value of Emp of 90 mV vs. SCE (Figure

2.20). It was observed that value of Emp for the zinc complex was lower than for the

ruthenium complexes A-C by approximate 30 mV, which clearly indicates that the

metal centre ion has an effect on the thermodynamics for the phendione ligand(s)

redox process. Similarly as for modifications with ruthenium complexes attached at

the surface through different linkers, minor differences were observed in the values

of Emp between electrodes 55-60. This result suggests that the length and structure of

linker does not affect the thermodynamics of redox reaction between complex D and

the electrode. Based on potential peak separation ∆Ep, the kinetics of electron

transfer are slower for zinc complex D than for ruthenium complexes an average

value of ∆Ep of 85 mV at a scan rate of 50 mV s-1 was calculated for electrodes 56-

61 and minor differences were observed between complex D attached at the surface

through the six different linkers.


80

-0.3 -0.2 -0.1 0.0 0.1 0.2

-8

-6

-4

-2

0

2

4

6

8

10

12

i /

µA

E vs. SCE / V

EDA

-0.3 -0.2 -0.1 0.0 0.1 0.2

-8

-6

-4

-2

0

2

4

6

8

10

12

BDA

i /

µA

E vs. SCE / V

-0.3 -0.2 -0.1 0.0 0.1 0.2

-8

-6

-4

-2

0

2

4

6

8

10

12

HDA

i /

µA

E vs. SCE / V

-0.3 -0.2 -0.1 0.0 0.1 0.2

-8

-6

-4

-2

0

2

4

6

8

10

12

XDA

i /

µA

E vs. SCE / V

-0.3 -0.2 -0.1 0.0 0.1 0.2

-8

-6

-4

-2

0

2

4

6

8

10

12

EDDAi /

µA

E vs. SCE / V

-0.3 -0.2 -0.1 0.0 0.1 0.2

-8

-6

-4

-2

0

2

4

6

8

10

12

BA

i /

µA

E vs. SCE / V

Figure 2.20 Cyclic voltammograms recorded for electrodes 56-61 modified with complex D attached

at the GC surface through six different linkers in presence (red) and absence (black) of 1 mM NADH

solution in 0.1 M phosphate buffer pH 7 at a scan rate of 50 mV s-1 and geometrical electrode area of

0.071 cm2.

In order to compare the efficiency of attachment of complex D through different

linkers, surface coverages for electrodes 56-61 were calculated using Equation 2.1.

In general, the surface coverage obtained for the series of modified electrodes with

the tetrahedral zinc complex D is higher than for electrodes modified with the more

bulky octahedral ruthenium complexes (Figure 2.21). Relatively high values of Γmed

were obtained for electrodes modified with the short aliphatic linkers EDA and BDA,

which is in good agreement with the coverages obtained for ruthenium complexes for

these linkers. The lowest values of Γmed were obtained for the modification with the

rigid benzylamine linker BA. Similar effects of the linkers on the values of Γmed were

observed for electrodes modified by the different ruthenium complexes.


81


0.02

0.04

0.06

0.08

0.10

0.12

complex D

Γm

ed /

10

-9 m

ol cm

-2

linker

Figure 2.21 Values of surface coverage Γmed of complex D calculated for modified electrodes 56-61

based on experimental data obtained from the cyclic voltammograms in Figure 2.15. Each bar

represents value of Γmed calculated according to Equation 2.2 for a single modified electrode.

Electrodes 56-61 were also individually evaluated towards NADH catalytic

activity in 1 mM NADH aqueous solution. It was found that phendione coordinated

to zinc metal ion catalyses oxidation of NADH at 0 V vs. SCE, which is in good

agreement with the results obtained for the ruthenium complexes. This result

indicates that the different metal centre and different geometry of complex does not

have an effect on the potential for catalysis of NADH oxidation. Catalytic currents,

icat, recorded at 0.15 V for electrodes 56-61 show minor differences in values of icat

between electrodes modified with diamine linkers. Relatively low values of icat were

recorded for electrode 61 attached through the benzylamine linker BA, which is

consistent with the low values of icat obtained for ruthenium complexes attached at

the surface through the rigid BA linker.

Further analysis involved calculations of normalised catalytic currents, inorm, using

Equation 2.2 to investigate the catalytic efficiency of electrodes 56-61 (Figure

2.22b). A significant value of inorm was obtained for modification with p-xylene

diamine linker XDA and a relatively low value of inorm was obtained for the rigid

benzylamine linker BA. In general, the calculated values of inorm for modifications

with zinc complex D are relatively low in comparison with values of inorm obtained

for modifications with the three ruthenium complexes A, B and C (Sections 2.3, 2.4

and 2.5).


82


0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0 complex Di c

at /

µA

linker

a)


7

14

21

28

35

42

complex D

i no

rm

/ 1

06 A

cm

5 m

ol-2

linker

b)



currents inorm calculated for electrodes 56-61 according to Equation 2.2. Each bar represents values of

icat (a) or inorm (b) obtained for single modified electrode.


83

2.7 XPS analysis of the GC electrodes modified with ruthenium

and zinc complexes.

The presence of complexes A, B C and D at the modified GC electrodes was

confirmed by X-ray photoelectron spectroscopy (XPS), which allowed for qualitative

and approximate quantitative determination of elements present at the GC surface

functionalised with ruthenium and zinc metal complexes. XPS method is based on

irradiation of atoms present in a solid sample with source of monochromatic soft X-

rays, with a photon energy range of 200-2000 eV, leading to ionisation and emission

of core shell electrons. These photons have limited penetrating power in a solid on

the order of 1-10 µm.151 The emitted electrons have measured kinetic energy Ekinet,

calculated using the photo-effect equation:

kinet bind= h - E Eν 2.3

where Ebind is the binding energy of the atomic orbital, from which the electron

originates and hν is the quantum energy of the photon emitted to the surface. In

practice, the instrument measures values of Ekinet at known hν and this allows

calculation of values of Ebind, which is sensitive characteristic of chemical bonds

present in the sample and is related to the energy of ionisation. Each element has a

range of electronic states which are open to excitation by x-rays, resulting in each

element having characteristic values of Ebind associated with electrons being ejected

from various orbital levels. The photoelectron spectrum is presented as a plot of

binding energy versus number of detected electrons per energy interval. Relative

intensities of the peaks correspond to energy released during ejection of electrons

from core orbital levels, which are characteristic for different atoms and type of a

bonding. This allows for quantitative evaluation of the sample composition and

character of bonding in compounds present at the solid surface.

In this work, XPS analysis was applied for determination of GC electrodes

covalently modified with ruthenium and zinc metal complexes according to the

strategy proposed in Scheme 2.2. Modifications of GC electrodes with ruthenium and

zinc complexes attached at the surface were reproduced at GC sheets using the same

electrochemical and solid-phase synthesis methods. Figure 2.23 shows three XPS


84

photoelectron spectra recorded for a blank GC surface and GC sheets modified with

ruthenium and zinc complexes and EDA linker. In all cases, broad peak at about 280

eV was observed, corresponding to electrons ejected from C 1s orbital level. High

intensity of this peak represents a large number of electrons ejected from sp2 carbon

atoms of the aromatic structures on the underlying GC sheets. In case of the modified

GC sheets (Figure 2.23b and c), the presence of carbon peaks corresponding to

electrons emitted from organic molecules occurred after functionalisation of GC was

not possible to distinguish due to high intensity of the background carbon atoms.151

In comparison with the blank GC (Figure 2.23a), increased intensity of electrons

ejection from the O 1s orbital level at 531 eV and the N 1s orbital level at 398 eV

were observed for functionalised GC surfaces suggesting successful attachment of

linker and complex (Figure 2.23b and c). However, relatively high intensity of the

oxygen peak might be caused by the presence of additional oxygen species directly

attached at GC surface as a result of electrochemical oxidation of the surface. For the

nitrogen peak, it was not possible to distinguish between single nitrogen-carbon bond

and the nitrogen-carbon double bond in the structure of bidentate ligands coordinated

to a ruthenium metal centre. The presence of the nitrogen with the same binding

energy was also observed at the blank GC, which might be rationalised by the

presence of residual nitrogen containing polymers used for production of GC

materials.52

Figure 2.23b shows characteristic peaks corresponding to emission of electrons from

ruthenium 3p1/2 and 3p3/2 orbital levels at values of Ebind of 484 and 462 eV,

respectively.151 According to the literature, ejection of electrons from ruthenium

3d3/4 and 3d5/2 orbital levels should occur at 284 and 280 eV. However, these peaks

overlap with the broad peak of C 1s and were not observed in this case. For GC sheet

modified with the zinc complex D, the XPS spectrum showed the presence of

additional two peaks at 1022 eV and 1020 eV corresponding to ejection of electrons

from Zn 2p3/2 and Zn 2p1/2 orbital levels, respectively.


85

1158 958 758 558 358 158

0

20000

40000

60000

Ekinet

/ eV

CP

S

Ekinet

/ eV

C 1s

N 1sO 1s

a)

1158 958 758 558 358 158

0

20000

40000

O 1s

C 1s

N 1sCP

S

Ru 3p1/2

Ru 3p3/2

b)

1121 921 721 521 321 121

0

5000

10000

15000

O 1s

N 1s

C 1s

CP

S

Ekinet

/ eV

Zn 2p3/2 Zn 2p1/2

c)

Figure 2.23 Examples of XPS photoelectron spectra recorded for GC sheets; a) polished, blank GC

sheet; b) with ruthenium complex A attached at the surface through EDA linker; c) zinc complex D

attached at the surface through EDA linker.

XPS analysis allowed also for approximate quantitative analysis of the elements

present at the modified GC sheets and comparison of the elemental composition

between different modifications. Firstly, each of the modifications with ruthenium

complexes A, B and C and zinc complex D attached through different linkers was

reproduced at GC sheets using optimised reaction conditions. In order to compare

elemental composition of the GC samples with different modifications, only two

elements were considered: nitrogen and metal ion. Carbon and oxygen elements were

omitted in the quantitative analysis due to presence of carbon background peak with

high intensity and possible additional oxygen species caused by oxidation of carbon

surface. Table 2.1 presents the theoretical atomic ratios of ruthenium and nitrogen

calculated for different modifications and experimental ratios of Ru : N obtained

from percentage atomic concentrations between different elements in samples

calculated using Casa XPS 2.3 software. The theoretical atomic ratios were based on

the assumption that each of the linker at the surface is attached to a metal complex.

Thus, we assumed that the theoretical atomic ratio for complexes A and D attached

through diamine linkers should be 1 : 6 and increased number of nitrogen per

ruthenium should be observed 1 : 8 for tris complexes B and C.

An approximate Ru/N ratio of 1 : 10 for the ruthenium complex A and the zinc

complex D was calculated from experimental data, which suggests that not all of the

linkers are coupled to the metal complex and a number of uncoordinated 2,2’-

bipyridine ligands is present at the GC surface. It might be rationalised by the bulky

size of the complex, which prevents coordination of each 2,2’-bipyridine ligand at


86

the surface to the large molecule of zinc or ruthenium complexes. For the tris

ruthenium complexes B and C, calculated approximate Ru/N ratios based on

experimental data were about 1 : 5 and 1 : 1 for complexes B and C, respectively.

The experimental results showed the decreased number of the nitrogen per one

ruthenium, which might be caused by the bulky size of complexes B and C and

prevents the photoelectron effect on nitrogen atoms on underlying linker structures.

Linker

Approximate atomic ratio

Complex A Complex B Complex C Complex D

Ru : N Ru : N* Ru : N Ru : N* Ru : N Ru : N* Zn : N Zn : N*

EDA 1 : 10 1 : 6 1 : 10 1 : 8 1 : 1 1 : 8 1 : 10 1 : 6

BDA 1 : 10 1 : 6 1 : 2 1 : 8 1 : 1 1 : 8 1 : 10 1 : 6

HDA 1 : 8 1 : 6 1 : 8 1 : 8 1 : 1 1 : 8 1 : 9 1 : 6

XDA 1 : 10 1 : 6 1 : 5 1 : 8 1 : 3 1 : 8 1 : 8 1 : 6

EDDA 1 : 7 1 : 6 1 : 4 1 : 8 1 : 1 1 : 8 1 : 7 1 : 6

BA 1 : 10 1 : 5 1 : 4 1 : 7 2 : 1 1 : 7 1 : 5 1 : 5

Table 2.1 The values of approximate atomic ratios Ru/N for the GC sheets modified with six different

linkers and four different metal complexes calculated from XPS experimental data (italics). Values of

Ru/N* represents theoretical atomic ratios Ru/N based on assumption that each linker at the surface is

coupled with a metal complex.

In conclusion, XPS analysis confirmed successful attachment of different

ruthenium and zinc complexes using the proposed strategy as shown in Scheme 2.2.

The XPS method allowed qualitative analysis of the modified GC sheets and the XPS

spectra clearly show peaks corresponding to the presence of ruthenium and zinc at

the modified GC sheets. XPS experimental data obtained for the modified GC sheets

were also applied for approximate quantification of elements Ru/N ratios in the

samples for bis ruthenium and zinc complexes and indicated the presence of

uncoordinated 2,2’-bipyridine ligands at the GC surface. However, presence of more

bulky complexes B and C at the GC sheets resulted in inaccurate ratios Ru/N, which

might be caused by limited photoelectron effect on the nitrogen atoms of the

underlying linkers.


87

2.8 Modification of GC electrode by novel zinc complex containing

pyridine imidazole ligands.

Modification of GC by octahedral ruthenium complexes A, B and C clearly

showed that presence of different ligands affects surface coverage of the attached

complex as well as catalytic activity towards NADH oxidation. Successful formation

of tetrahedral zinc complex D at the GC surface was achieved by solid-phase

coordination of complex 52 by 2,2’-bipyridine ligand covalently immobilised at GC

through different linkers (Section 2.3). In order to study the effect of different ligands

on the zinc complex, chelating ligands containing the imidazole heterocyclic ring

were considered. Zinc complexes with coordinated bidentate and tridentate imidazole

ligands are reported in the literature and their interesting chemical and structural

properties have been applied in materials science.152-154 Chelating ligands with

imidazole structures were also reported by Schuhmann et al. 155 for solid-phase

functionalisation of GC electrodes by osmium complexes. In this case, the pyridine-

imidazole ligand was functionalised at one of the imidazole nitrogen atoms by a

hydroxyethylene group and covalently tethered at the polymer supported GC using

solid-phase reaction conditions. In the presence of bis osmium complexes, the

pyridine-imidazole coordinated to the osmium metal centre and created a new, stable

osmium complex at the GC surface, which was applied for bioelectrocatalytic

oxygen reduction.

In this project, the design of a functionalised pyridine-imidazole with a carboxylic

functional group was proposed, which would allow for covalent tethering of this

ligand at GC and enable formation of the zinc complex. Based on the synthetic

approach reported by Schuhmann et al.,155 the pyridine-imidazole was functionalised

at imidazole N-1 in a three steps synthesis and 2-(2-(pyridine-2-yl)-1H-imidazol-1-

yl)acetic acid 64 was successfully obtained as the TFA salt form (Scheme 2.18).

Firstly, 2-(1H-imidazol-2-yl)pyridine 61 was obtained using a literature

procedure156,157 in 31 % yield (lit.157 33 %). Addition of 2-bromo-tert-butyl acetate to

a solution of 62 in the presence of n-butylithium allowed nucleophilic substitution of

the bromide by nitrogen in the imidazole ring at N-1 position to give compound 63 in

76 % yield. Structure of 63 was confirmed by NMR analysis. 1H NMR shows the


88

presence of resonance peaks in the aromatic region corresponding to pyridine and

imidazole heterocyclic rings. In addition, a singlet at 5.2 for two α-protons of acetate

group and a singlet at 1.39 ppm for tert-butyl ester were observed. Finally,

hydrolysis of the ter-tbutyl ester 63 in 20 % solution of TFA gave ligand 63 in 98 %

yields.

Scheme 2.18 Synthesis of ligand 64; a) 2.5 M n-buthylithium in hexane, 2-bromo-tertbutyl acetate

THF, 78 °C, overnight, 76 %; b) 20 % trifluoroacetic acid (TFA) in DCM, overnight, R. T., 98 %.

The following work involved covalent immobilisation of ligand 64 under

solid-phase peptide coupling conditions as reported for attachment of 2,2’-bipyridine

ligand 9 (Scheme 2.19). In this case, diisopropylethylamine (DIPEA) base was used

in excess of 10 equivalents in order to neutralise the TFA salt. The presence of the

attached pyridine-imidazole ligand at electrode 65 was difficult to evaluate directly

using cyclic voltammetry due to the relatively low potential of the imine redox

couple as discussed for electrode 10 (Section 2.3). Therefore, electrode 65 was

directly used in the following step of coordination of the zinc complex 53 under

solid-phase reaction conditions as optimised for coordination of the 2,2’-bipyridine

to 53.

Scheme 2.19 Preparation of modified electrode 66; a) 1 M solution of 60 in DMF, 10 equiv. of

DIPEA, HBTU, R.T., 16 h; b) 0.01 M solution of 53 in DMF, 50 °C, 16 h.

In order to confirm the successful coupling of the novel ligand 64 to the amine

groups at the GC, a control experiment was performed. Attachment of 53 was carried

out at the same time by dipping electrode 10 containing mono-Boc-EDA linker and

electrode 65 modified with pyridine-imidazole ligand. Results of the control

experiment were evaluated by cyclic voltammetry. For both modified electrodes, a


89

reversible redox peak corresponding to a redox process of the phendione was

observed at Emp of −0.1 V vs. SCE (Figure 2.24). However, electrode 66 gave

significant coverage Γmed of 1.4 × 10−10 mol cm-2 in comparison to the control

electrode with a value of Γmed 1.23 × 10−12 mol cm-2.

-0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2

-6

-4

-2

0

2

4

6

electrode 62

control electrode

i /

µA

E vs. SCE / V

Figure 2.24 Cyclic voltammograms of electrode 66 (black) and control electrode (red) in 0.1 M

solution of phosphate buffer pH 7 at a scan rate 50 mV s-1 and geometrical electrode area of 0.071

cm2.

In order to evaluate the effect of the pyridine-imidazole on the catalytic activity for

NADH oxidation, electrode 66 was characterised in 0.1 M solution of NADH and

shows increased, irreversible anodic process at 0 V vs. SCE (Figure 2.25). The

catalytic current, icat, was measured at 0.15 V vs. SCE and is comparable with icat

obtained for complex D. Functionalisation of the GC surface by complex E attached

at the surface through different linkers followed by their detailed electrochemical and

kinetic characterisation is included in the library of 63 modified electrodes described

in Section 3.


90

-0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2

-6

-4

-2

0

2

4

6

i /

µA

E vs. SCE / V

Figure 2.25 Cyclic voltammograms of electrode 66 in the presence (red) and absence (black) of 1 mM

NADH solution in 0.1 M phosphate buffer pH 7 at scan rate of 50 mV s-1 and geometrical electrode

area of 0.071 cm2.


91

2.9 Conclusion

A strategy for functionalisation of GC electrodes was designed in order to covalently

immobilise metal complexes containing phendione redox active ligands for NADH-

dependent biosensors. The strategy, based on sequential electrochemical and solid-

phase synthesis steps, allows for control over choice of metal centre, ligand and

linker.

According to the proposed strategy, the initial step involved electrochemical

oxidation of diamine or reduction of diazonium salt bearing amine functional groups.

The presence of amine groups at the GC surface allowed for successful covalent

attachment of the 2,2’-bipyridine-5-carboxylic acid bidentate ligand under solid-

phase peptide coupling conditions. The final step involved coordination of 2,2’-

bipyridine to a metal centre and formation of a novel complex at the surface.

Initially, octahedral bis-(dimethyl sulphoxide)(1,10-phenanthroline-5,6-dione)

ruthenium (II) chloride was successfully attached at the surface through

ethylenediamine linker under optimised solid-phase reaction conditions to create the

novel bis ruthenium complex A. Electrochemical characterisation of the modified

electrode showed that complex A exhibits reversible redox behaviour for the

phendione and acts as an effective mediator towards NADH oxidation.

Based on the initial result, the same strategy was applied for attachment of octahedral

ruthenium complexes, containing various chelating ligands. For this reason, the

additional ruthenium complexes were synthesised: (2,2’-bipyridine)(1,10-

phenanthroline-5,6-dione) ruthenium (II) chloride and bis-(1,10-phenanthroline-5,6-

dione)ruthenium (II) chloride and coordinated to 2,2’-bipyridine at the GC surface

under optimised solid-phase reaction conditions, resulting in formation of novel

ruthenium complexes B and C, respectively. In order to study effect of the geometry

of the metal complex, the tetrahedral (1,10-phenanthroline-5,6-dione)zinc (II)

chloride complex was attached at the GC surface according to the proposed strategy

to give the novel tetrahedral complex D. Results of electrochemical characterisation

showed that all of the electrodes were successfully modified with complexes B-D

showing semi-reversible redox behaviour for the phendione. Surface coverages were

calculated for the different ruthenium complexes, indicating that the value of

coverage strongly depends on the size of the metal complex at the surface, where


92

relatively low coverage was obtained for attachment of the bulky ruthenium

complexes B and C and significant coverage was obtained for modification with the

tetrahedral zinc complex D.

Different metal complexes were evaluated towards NADH oxidation and preliminary

studies showed electrocatalytic activity at low overpotential in all cases. Significant

catalytic efficiency was obtained for ruthenium complexes B and C in comparison to

the zinc complex D, which would suggest that octahedral complexes with additional

bidentate ligands increase the catalytic activity of the modified electrodes for NADH

oxidation. This might be explained by the fact that delocalised Π-electrons in the

2,2’-bipyridine ligand, which increases the electron transfer between phendione and

the electrode surface.

Further work involved preparation of electrodes modified with complexes A, B

and C attached at the surface through various linkers in order to study the effect of

length and structure of the linker on the electrochemical and catalytic properties of

the complexes. In general, values of surface coverages calculated for ruthenium

complexes A-D decrease with long aliphatic linkers. Relatively low coverages were

observed for the benzylamine linker, where the direct bond between the benzyl ring

and the carbon surface forms a rigid structure at the surface, resulting in limited

accessibility of the 2,2’-bipyridine to coordinate to the bulky ruthenium complex.

These results are with good agreement with the results reported for immobilisation of

the anthraquinone using the same methodology.

Attachment of complexes A-D through linkers with different flexibility and length

also affected catalytic activity of the modified electrodes for NADH oxidation. In

general, significant catalytic efficiency was obtained for flexible EDDA and XDA

linkers for all complexes. Relatively low catalytic efficiency was calculated for the

short aliphatic linkers EDA and BDA and the rigid benzylamine linker BA, which

can be rationalised by low accessibility of these linkers to the bulky, folded molecule

of NADH in solution.

The presence of complexes A-D attached at the surface through different linkers was

confirmed by X-ray photoelectron spectroscopy, which allowed for qualitative and

quantitative determination of elemental composition of the modified electrode

surface. Comparison of XPS spectra for the blank GC surface and GC functionalised


93

by complexes A-D according to the proposed strategy, showed the presence of peaks

characteristic for ruthenium or zinc elements.

In order to investigate the effect of different ligands on the coverage and

catalytic activity of the tetrahedral zinc complex, the novel 2-(2-(pyridine-2-yl)-1H-

imidazol-1-yl)acetic acid chelating ligand was synthesised and attached to the

surface through an ethylenediamine linker according to the proposed strategy. The

(1,10-phenanthroline-5,6-dione)zinc (II) chloride complex was successfully

coordinated by the pyridine-imidazole at the GC surface and the novel, tetrahedral

zinc complex E was successfully formed under solid-phase reaction conditions.

Electrochemical characterisation of complex D at the GC surface indicates that this

complex acts as an effective mediating system for NADH oxidation. Attachment of

complex E through different linkers and detailed electrochemical characterisation are

included in Section 3 describing preparation and HTP evaluation of a library of

different modified electrodes.

In conclusion, molecular design of GC electrodes for electrocatalysis of NADH

oxidation was successfully achieved by applying electrochemical and solid-phase

methods. The proposed strategy involved sequential functionalisation of GC

electrodes in a controlled manner, designed for covalent tethering of different metal

complexes bearing the phendione redox active ligand. As a result, metal complexes

with different geometry and ligand(s) were attached at the surface through linkers

with various lengths and flexibility and successfully applied as an effective

mediating system for oxidation of NADH. Initial studies of the individual modified

electrodes clearly showed that the proposed design of the electrode surface allowed

for control over electrocatalytic performance of the modified electrodes. Synthesis

and electrochemical results presented in this chapter was applied for more detailed

investigation of modifications A-E and comparison with the literature examples in

Chapter 3.


94

Library of modified electrodes - preparation and HTP screening

95

3 Library of modified electrodes - preparation and HTP screening

3.1 Objectives

In the previous chapters, methodology and optimisation for the covalent

modification of GC surfaces by various metal complexes were described using the

strategy outlined in Scheme 3.1.

Scheme 3.1 General approach to covalent modification of GC electrodes by metal complexes.

As described in Section 2, ruthenium (II) and zinc (II) metal complexes with

phendione bidentate ligand(s) were successfully attached to a GC surface through

different linkers. The single electrode results show that the nature of the linker and

structure of the complex have an impact on the amount of attached metal complex at

the surface and its catalytic activity of towards NADH oxidation.

Applying the same methodology, we therefore set out to prepare a library of

modified electrodes, which could be electrochemically characterised in HTP mode

using parallel cyclic voltammetry. It was anticipated that HTP would allow the direct

and rapid comparison of many different modifications during a single set of

measurements and rapid identification of the best modification with respect to the

surface coverage and catalytic activity towards NADH oxidation.

Twenty different modifications were planned with each repeated three times to assess

the variability of different modifications towards NADH catalytic oxidation. The

final three members of the library were control electrodes. The design of the library

is presented in Table 3.1. The four selected linkers offer a variety of functional


96

groups and lengths in order to investigate their affect on the surface coverage of the

attached metal complex and catalytic activity towards NADH oxidation. Two of the

linkers have aliphatic chains with different lengths (EDA and HDA). The third was a

flexible 2,2'-(ethylenedioxy)diethylamine chain (EDDA), in which hydrophobic

aliphatic chains are replaced by more hydrophilic ethoxy units. Benzylamine BA was

also selected as a linker, in which the aromatic ring was directly bonded to the GC

surface to create a short and rigid linkage between mediator and electrode surface

(Table 3.1).

Single electrode experiments had shown that bulky octahedral ruthenium complexes

A, B, C and tetrahedral zinc complex D when attached through different linkers at

the surface gave different catalytic activities towards NADH oxidation. The same

metal complexes were selected for the library in order to investigate effect of size

and geometry of the metal complex on catalytic activity towards NADH catalytic

oxidation (Table 3.1). In addition, three octahedral ruthenium complexes A, B and C,

which were different in the size and structure of the chelating ligands attached to

ruthenium metal ion. These included chloride ions (complex A), 2,2’-bipyridine

(complex B) and phendione ligand (complex C). This allowed us to observe possible

effect of these ligands on the electrochemical and catalytic properties of the

complexes A, B and C attached at the surface.

Successful modification of electrodes by complex E attached at the surface through

the EDA linker has been described previously in Section 2.8. The library allowed us

to explore this modification by including complex E attached through the four

different linkers.

Electrodes 70a-c (Table 3.1) were modified with the EDA linker, the 2,2’-bipyridine

ligand at the surface and the electrochemically inactive (2,2’-bipyridine)zinc (II)

chloride are included as a control modification to observe the electrochemical

changes due to covalent functionalisation by a metal complex in the absence of the

redox active phendione ligand.


97

EDA HDA EDDA BA

A 13a, 13b, 13c 36a, 36b, 36c 37a, 37b, 37c 39a, 39b, 39c

B

41a, 41b, 41c 43a, 43b, 43c 44a, 44b, 44c 46a, 46b, 46c

C

47a, 47b, 47c 49a, 49b, 49c 50a, 50b, 50c 52a, 52b, 52c

D

56a, 56b, 56c, 58a, 58b, 58c 59a, 59b, 59c 61a, 61b, 61c

E 66a, 66b, 66c 67a, 67b, 67c 68a, 68b, 68c 69a, 69b, 69c

control electrodes

70a, 70b, 70c

Table 3.1 Design of the library of 63 modified electrodes; the columns represent the four linkers:

EDA (1,2-ethylenediamine), HDA (1,6-hexanediamine), EDDA (2,2'-(ethylenedioxy)diethylamine)

and BA (p-benzylamine); the rows represent: the ruthenium complexes (A, B and C), the zinc

complexes (D, E) and the control electrodes 70a-c.

N N

O

RuN

NO

O

N

N

2Cl-linker

+2


98


99

3.2 Preparation and synthesis of the 63 modified electrode library.

Preparation of the 63 member library started with individual polishing of the

electrodes followed by electrochemical covalent attachment of four linkers at

electrode surface carried out one by one using conventional cyclic voltammetry.

As was observed for individually modified electrodes (Section 2), attachment of the

linker at the electrode surface generates currents in range of 200-300 µA, whereas

the HTP multichannel potentiostat has a current limit of 50 µA. Therefore,

attachment of the linkers for the 63 member library was not possible to perform in a

parallel way using the multichannel potentiostat. However, subsequent steps of

electrode modification involving solid-phase synthesis were carried out in a

combinatorial and parallel way.

Firstly, the electrodes were polished with silicon carbide paper grade P1200

prior to covalent modification. High roughness or defect levels obtained during the

polishing process might cause changes in the effective surface areas of the electrodes

and be a source of variations between electrodes with the same modification.

Polished blank electrodes were therefore screened one by one using cyclic

voltammetry in phosphate buffer solution at pH 7. The resulting voltammograms

gave consistent capacitance nonfaradaic currents (Figure 3.1). Comparison of the

background currents for the blank electrodes gives an approximate view of the

differences in their active areas after the polishing process. Voltammograms of all

polished electrodes are presented in Appendix 1 and they clearly show that the

capacitance currents measured at potential of 0 V vs. SCE is consistent for all of the

polished blank electrodes.


100

-0.2 -0.1 0.0 0.1 0.2

-3

-2

-1

0

1

2

3

i /

µA

E vs. SCE/V

electrode 13a

electrode 13b

electrode 13c

Figure 3.1 Typical cyclic voltammograms of background currents recorded for blank glassy carbon

electrodes 13a-c with geometrical area of 0.71 cm2 in 0.1M phosphate buffer solution pH 7 at scan

rate of 50 mV s-1.

The first step of covalent modification was attachment of the mono-Boc

protected linker as previously reported for the modification of individual electrodes

(Section 2).55 For attachment of the diamine linkers EDA, HDA and EDDA, the

oxidation peak on the first cycle was observed at a potential of 1.96 V (Figure 3.2a),

which is in good agreement with the results previously obtained for individual

electrodes. The following cycles led to the disappearance of the faradaic curve,

which indicates blocking of the surface by the Boc-protected amine functional

groups and creates monolayer of the diamine linker at the GC surface. Similarly,

during the attachment of the diazonium salt linker (BA), the reduction wave at −0.36

V was observed on the first cycle and decreased in the following cycles, which

indicates blocking of the surface with the BA linker (Figure 3.2b). The electrode was

covered with an expected monolayer of the BA linker after two cycles.


101

0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4

-50

0

50

100

150

200

250

300

i /

µA

E vs. SCE / V

a)

-1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8

-8

-6

-4

-2

0

2

i /

µA

E vs. SCE / E

b)

Figure 3.2 Typical cyclic voltammograms for linkers immobilisation at GC polished electrodes in 15

mM solution of linker in MeCN with 0.1 M TABTFB at a scan rate of 50 mV s-1; a) attachment of the

EDA linker for electrode 13c; b) attchment of the BA linker for electrode 39a.

The further steps of modification involved parallel solid-phase synthesis.

Firstly, the Boc amine protecting group was removed by dipping all the electrodes in

a bulk solution of 0.1 M hydrochloric acid in dioxane for 1 h. This time was enough

to complete the deprotection reaction as shown in the previous results obtained for

modifications of the individual electrodes (Section 2) and based on results reported

Kilburn and Bartlett for electrodes modified with anthraquinone-2-carboxylic

acid.55,57

The resulting free amine groups allowed for the attachment of 2,2’-bipyridine-5-

carboxylic acid bidentate ligand 9 using peptide coupling conditions previously

reported in the literature (Scheme 3.2).55,136 The coupling reaction of ligand 1 was

assumed to be completed within the same time (16 h) as previously shown in

modifications of the individual electrodes with the ligand 9 (Section 2.3). In order to

obtain identical coupling conditions during library preparation and decrease the

experimental errors associated with the coupling reaction, the following electrodes

13a-c, 41a-c, 47a-c, 56a-c, 36a-c, 43a-c, 49a-c, 58a-c, 37a-c, 44a-c, 50a-c, 59a-c,

39a-c, 46a-c, 52a-c, 61a-c were dipped in one bulk solution of reaction mixture for

exactly the same amount of time.

The preliminary experiments for attachment of the 2-(1-(carboxymethyl)-1H-

imidazol-3-ium-2-yl)pyridin-1-ium trifluoroacetate salt 64 clearly showed that

successful attachment of this linker was obtained in similar conditions as for linker 9

but with 10 equiv. excess of the DIEA base in order to neutralise the TFA salt. As


102

shown for ligand 9, the coupling of ligand 64 was carried out in a parallel way,

where the following electrodes 66a-c, 67a-c, 68a-c and 69a-c were all dipped in the

one bulk solution of the reaction mixture for exactly the same amount of time.

Scheme 3.2 Attachment of linkers 9 and 64 at the functionalised GC surface; a) ligand 9, HBTU,

DIPEA, DMF, R.T., 16 h; b) ligand 64, HBTU, 10 equiv. DIEA, DMF, R.T., 16 h.

The presence of the bidentate ligand allowed for attachment of the ruthenium (II)

metal complexes A, B and C or the zinc (II) metal complexes D and E by applying

the conditions of solid-phase synthesis optimised in Section 2 (Scheme 3.3).


103

Scheme 3.3 Solid-phase synthesis of metal complexes at the functionalised GC electrodes; a)

Ru(phendione)Cl2(DMSO)2, DMF, 100 °C, 16 h. b) Ru(phendione)(bipy)Cl2, DMF, 100 °C, 16 h, c)

Ru(phendione)2Cl2, DMF, 100 °C, 16 h, d) and e) Zn(phendione)Cl2, DMF, 50 °C.

The solid-phase synthesis was carried out in parallel, using four-neck vessels, where

each vessel contains four electrodes dipped in the solution of the appropriate metal

complex (Figure 3.3).


104

Figure 3.3 Example of the parallel set up of reaction vessels used during solid-phase synthesis of

metal complexes at GC electrodes; the inset shows the reaction vessel containing a solution of

Zn(phendione)Cl2 and electrodes 56c, 58c, 59c and 61c.

Table 3.2 shows the arrangement of the electrodes in vessels I to XVI. The electrodes

modified with the same linker and bidentate ligand were placed in three different

vessels, containing the same solutions of metal complex. The parallel preparation of

the electrodes with the same modification but using different vessels allows us to

evaluate the reproducibility and variation in each modification. Parallel and

combinatorial preparation of all members of the library allowed us to avoid possible

problems associated with low stability of the modified electrodes.


105

Metal complex

Number of

reaction

vessel

Linker/Electrode number

EDA HDA EDDA BA

A I 13a 36a 37a 39a

II 13b 36b 37b 39b

III 13c 36c 37c 39c

B IV 41a 43a 44a 52

V 41b 43b 44b 53

VI 41c 43c 44c 54

C VII 47a 49a 50a 52a

VIII 47b 49b 50b 52b

IX 47c 49c 50c 52c

D

X 56a 58a 59a 61a

XI 56b 58b 59b 61b

XII 56c 58c 59c 62c

E

XIII 66a 67a 68a 69a

XIV 66b 67b 68b 69b

XV 66c 67c 68c 69c

Control electrodes

XVI 70a-c

Table 3.2 Arrangment of electrodes and reaction vessels during parralel attachment of the metal

complexes.


106


107

3.3 HTP electrochemical screening of the library.

The library of 63 modified electrodes was electrochemically screened in HTP

mode using a multichannel potentiostat. Firstly, the electrodes were cycled in 0.1 M

phosphate buffer solution, over the range −0.3 to 0.2 V to observe the surface redox

process for the phendione ligand coordinated to the ruthenium or zinc metal ions

(Appendix 3). The effect of sweep rate was studied by screening the library at range

of scan rates from 10 to 500 mV s−1. The anodic and cathodic peak currents increase

linearly with scan rate as shown in Figure 3.4. This is consistent with surface

immobilisation of the metal complexes and was observed for all modified electrodes

in the library.55,57

-0.3 -0.2 -0.1 0.0 0.1 0.2

-40

-20

0

20

40

10 mV/s

20 mV/s

50 mV/s

100 mV/s

200 mV/s

500 mV/s

i med / µ

A

E vs. SCE/V

a)

0 100 200 300 400 500

-40

-30

-20

-10

0

10

20

30

40

i med /

µA

ν / mVs-1

anodic currents

cathodic currents

b)

Figure 3.4 a) Effect of scan rate on modified GC electrode 13a in 0.1 M phosphate buffer solution pH

7, electrode area of 0.71 cm2; b) Plot of anodic (red) and cathodic (black) currents as a function of

scan rate ν for modified electrode 13a.

Table 3.3 reports the anodic (Ea) and cathodic (Ec) peak potentials, peak separation

(∆Ep) and midpoint potential (Emp) (Figure 3.5). In general, the choice of the metal

ion and geometry of the metal complex affect the kinetics of electron transfer as

shown by the difference between anodic and cathodic potentials for the tetrahedral

zinc and the octahedral ruthenium complexes.


108

Linker Metal

ion

Complex Ea vs.

SCE/mV

Ec vs.

SCE/mV

∆∆∆∆Ep vs.

SCE/mV

Emp vs.

SCE/mV

EDA Ru A −33 −78 −45 −56

Ru B −33 −86 −53 −60

Ru C −39 −84 −45 −61

Zn D −44 −137 −93 −91

Zn E −47 −135 −88 −91

HDA Ru A −23 −81 −58 −52

Ru B −17 −82 −65 −50

Ru C −33 −82 −49 −58

Zn D −35 −143 −108 −89

Zn E −30 −151 −121 −90

EDDA Ru A −24 −75 −51 −50

Ru B −20 −87 −67 −54

Ru C −34 −84 −50 −59

Zn D −34 −147 −113 −91

Zn E −12 −144 −132 −77

BA Ru A −15 −98 −83 −56

Ru B −10 −93 −84 −52

Ru C −20 −92 −72 −56

Zn D −5 −144 −139 −75

Zn E −35 −147 −112 −91

Table 3.3 Summary of anodic and cathodic potentials peaks Ea and Ec, separation peak ∆Ep, midpoint

potential Emp = (Ea + Ec) / 2 obtained from cyclic voltammograms recorded in 0.1 M phosphate buffer

solution pH 7 at a scan rate of 50 mV s−1.

Slower electron transfer was observed for electrodes modified with zinc complexes

where the average ∆Ep of 113±6 mV is significantly greater than for a ruthenium

modified electrodes where ∆Ep of 60±4 mV. The choice of the metal ion also affects


109

the thermodynamics of the phendione ligand redox process as shown by the variation

in the midpotential Emp values with average Emp of −55±1 mV vs. SCE for the

ruthenium complexes and Emp of −86±2 mV vs. SCE for the zinc complexes.

Similar values of ∆Ep and Emp were observed between the metal complexes A, B and

C, which might suggest that the presence of various ligands such as chloride ions,

2,2’-bipyridine ligand or increased number of the phendione ligands (complex C),

have negligible effect on the kinetics and thermodynamics of the modified electrodes

(Figure 3.5). A similar situation was observed for the electrodes modified with

tetrahedral zinc complexes D and E, where the presence of different chelating ligands

did not affect the values of ∆Ep and Emp. Previously reported modification of the GC

electrode by various linkers and anthraquinone as the redox probes showed that the

nature of the linker has an effect on the electron transfer kinetics. The values of ∆Ep

were found to increase with increasing chain length of the aliphatic linkers.

Relatively high values of ∆Ep were observed for electrodes functionalised with the

benzylamine linker BA.55,57 A similar effect of the linker on the electron transfer

kinetics was observed between different linkers in the library.

Based on the peak separation data ∆Ep, the electron transfer kinetics are faster for

electrodes modified with the aliphatic linkers EDA, HDA and EDDA rather than the

benzylamine linker BA. In the case of the ruthenium modified electrodes average

values of ∆Ep of 54±3 and ∆Ep of 80±4 mV were obtained for aliphatic and BA

linkers, respectively. The values of the peak separation ∆Ep for the modified

electrodes with the zinc complexes were relatively smaller with ∆Ep of 109±7 mV

for the aliphatic linkers and ∆Ep of 125±14 mV for the BA linker. However, the type

of linkage does not influence the redox potential (Emp) of the modified electrodes.


110

EDA HDA EDDA BA

-140

-120

-100

-80

-60

-40

-20

0 complex A

complex B

complex C

complex D

complex E

∆E

p / m

V

linker

a)

EDA HDA EDDA BA

-90

-80

-70

-60

-50

-40 complex A

complex B

complex C

complex D

complex E

Em

p / m

V

linker

b)

Figure 3.5 Average values of the separation pontential Ep (plot a) and the midpoint potential Emp (plot

b) observed for twenty different modifications in the library based on the experimental data recorded

in 0.1 M phosphate buffer solution pH 7 at scan rate 50 mV s−1.

The efficiency of the different modifications for the 63 member library was

determined by calculating the amount of immobilised metal complex with redox

active phendione ligands. The surface coverage, Γmed, of the modified electrodes was

calculated according to Equation 2.1. Figure 3.6a shows the summary of the mean

values of the surface coverage Γmed (mol cm−2) calculated for twenty different

modifications in the library and the standard errors of the mean value associated with

each modification. The average value of Γmed for each modification was calculated

from three replicate electrodes. The reproducibility of each modification was

justified by calculations of the standard errors of the mean using Origin 8.1 software.

Electrodes modified with different linkers and metal complexes show variations in

the calculated values of the coverages Γmed. The coverage is related to the size of the

immobilised metal complex at the surface. Electrodes modified with the relatively

smaller zinc (II) complexes D and E show significantly higher coverage than

electrodes modified with the bulkier ruthenium (II) complexes. The coordination

environment also affects the surface coverage as observed for the electrodes

modified by the three different ruthenium complexes A, B and C. Higher coverages

were observed for complex A, which contains two relatively small chloride ligands.

Substitution of the chloride by the more bulky 2,2’-bipyridine or phendione bidentate

ligands decreases the coverage for the electrodes modified with complexes B and C.

The presence of different bidentate ligands at the surface does not affect the coverage

as shown by comparison between the modifications with the complexes D and E.


111

The attachment of linkers with different functionalities and chain lengths also

influences the values of the surface coverage of the metal complexes. Electrodes

functionalised with the EDA linker show higher values of Γmed in comparison with

electrodes modified with the longer aliphatic linkers HDA and EDDA. These results

are in good agreement with modifications of the individual electrodes (Section 2).

This might be rationalised by the fact that the coverage obtained after initial

electrochemical attachment of the mono-Boc protected diamine linkers decreases for

the amines with longer aliphatic chains. Approximate values of the surface coverages

of the diamine linkers with different chain lengths were calculated based on XPS

experimental data and clearly show that the coverages of the diamine linkers

decrease with the length of the aliphatic chain.57 The presence of the longer aliphatic

chain in the diamine might cause a conformational disorder within the alkyl chains

and hinder the access of the free amine group to the carbon surface thus preventing

the amine from covalent attachment to the GC surface.149 As a result, the surface

coverages calculated from XPS data for the longer linkers are relatively smaller than

for electrodes modified with shorter linkers such as EDA. Thus, lower number of the

linker at the GC decreases yields of the subsequent steps of the electrode

modification (Scheme 3.3), resulting in lower final coverage of the redox active

metal complex attached at the surface.

Relatively low values of Γmed were also obtained for electrodes modified by the BA

linker. In this case, it might be explained by the limited flexibility of the bond

between the aromatic ring of the linker and the GC surface. The rigid structure of the

linker at the surface might prevent the 2,2’-bipyridine ligand from coordinating to the

metal ion in the relatively large metal complex.

These results show that the length and flexibility of the linkers and the geometric

arrangement of the metal complex at the surface all affect the surface coverage for

the different modifications.


112

Figure 3.6 Average values of the surface coverage Γmed of the metal complex calculated for 20

different modifications in the library using Equation 2.1 based on experimental data recorded in 0.1 M

phosphate buffer solution pH 7 at scan rate of 50 mV s−1; geometrical electrode area of 0.071 cm2;

Complexes A, B, D and E contain one phendione ligand whereas complex C has two phendione

ligands; a) Each bar represents mean values of Γmed with the and standard error of the mean for three

replicate electrodes; b) three dimensional representation of the means.

BAEDDA

HDAEDA

2

4

6

8

10

A

B

C

D

E

Γ /

10

-11 m

ol cm

-2

linker

complex

b)

EDA HDA EDDA BA0

2

4

6

8

10

12 complex A

complex B

complex C

complex D

complex E

Γ /

10

-11 m

ol cm

-2

linker

a)


113

3.4 Catalytic oxidation of NADH

The electrochemical HTP screening of the library was applied for rapid evaluation

of the catalytic activity of the library towards NADH catalytic oxidation in a single

set of parallel measurements.

The initial experiment was performed in a freshly prepared 1 mM aqueous solution

of NADH at 50 mV s−1. Figure 3.7 compares cyclic voltammograms in the absence

(red line) and presence (black line) of NADH for modified electrode 13a. The results

are typical for all the modified electrodes in the array. A decrease in the potential of

NADH oxidation ENADH to 13±1 mV vs. SCE was observed relative to oxidation of

NADH at bare GC, clearly indicating an electrocatalytic effect for all of the

modifications.158 A full set of cyclic voltammograms in the presence and absence of

NADH for all members of the library are given in Appendix 2.

-0.3 -0.2 -0.1 0.0 0.1 0.2

-10

-5

0

5

10

i /

µA

E vs. SCE / V

1mM NADH

mediator

Figure 3.7 An example of a cyclic voltammograms recorded in the presence (black) and absence (red)

of 1 mM NADH in 0.1 M phosphate buffer solution pH 7 at a scan rate of 50 mV s−1, geometrical

electrode area of 0.071 cm2. Data shown in Figure 3.7 corresponds to a modified electrode 13a.

The catalytic currents, icat, were defined as the difference between the anodic

currents in the presence and absence of NADH at a potential of 0.15 V. Figure 3.8

shows a summary of average values of icat obtained for twenty different

modifications in the library; the associated standard errors of the means for all of the

modifications does not exceed twenty percent. This error might be related to the

differences in roughness between individual electrodes. Each mean value of icat

represents currents in the presence of NADH recorded for three replicate electrodes.


114

BAEDDA

HDAEDA

0.0

0.7

1.4

2.1

2.8

3.5

4.2

A

B

C

D

E

i cat / µ

A

linker

complex

b)

In general, the trends in the values of icat reflect the surface coverage of the metal

complexes for electrodes modified with the ruthenium complexes A, B and C, where

the values of icat increases with higher coverage of the metal complex (Figure 3.6). In

this case, increase of the values of icat with the surface coverage would suggest all the

phendione ligands present at the surface participate in the catalytic reaction.

The values of icat obtained for the modifications with the zinc complexes D and E are

lower than would be expected for relatively high coverages calculated for the

modifications with these complexes. This might suggest that not all of the phendione

ligands coordinated to the zinc metal ion participate in the oxidation of the NADH.

This explanation could be confirmed by designing modified GC electrodes, where

zinc complexes D and E would create less than a full monolayer of the complex at

the GC, which could be obtained by introduction of different blocking (capping)

groups. The partial surface coverage of complexes D and E would be then studied for

dependence of the coverage on the NADH catalytic currents.

The measured values of icat obtained for the various linkers are also affected by the

surface coverage (Figure 3.8). The most pronounced catalytic activity was obtained

for modifications with the EDA linker and decreases with longer aliphatic linkers as

shown in modifications with HDA and EDDA linkers. Relatively low values of icat

were found for electrodes modified with the BA linker, presumably due to its limited

flexibility, which might prevent the orientation of the phendione ligand necessary for

hydride transfer between the quinone groups at the surface and NADH molecules.

EDA HDA EDDA BA0

1

2

3

4

5

ica

t / µ

A

linker

complex A

complex B

complex C

complex D

complex E

a)

Figure 3.8 a) Means of the catalytic currents icat for 20 different modifications of the library measured

in 1 mM NADH solution at a scan rate of 50 mV s−1. Each bar represents an average value of icat and


115

the standard errors of the means calculated for 3 replicate electrodes; b) The means of icat values for 20

different modifications of the library presented in three dimensional bar plot.

In order to justify the differences in the catalytic efficiency between various

modifications, the catalytic currents were normalised according to Equation 2.2.

The values of catalytic currents obtained for 1 mM NADH and presented in Figure

3.8 were initially normalised by surface coverage of the phendione ligands (Figure

3.6).

In general, the structure and length of the linkers has a pronounced effect on the

values of inorm (units of A cm5 mol−2) (Figure 3.9). The values of inorm calculated for

electrodes modified with the aliphatic linkers show higher values of inorm than for

electrodes modified with the BA linker, which might be related to the limited

flexibility of this linker as discussed previously. On the other hand, structure and

geometry of the metal complex has a smaller effect on the values of inorm. However,

the general trend was observed for complexes B and C with phendione and 2,2’-

bipyridine ligands showing slightly higher catalytic activity than complexes A, D or

E. This might suggest that the presence of additional ligands with conjugated

aromatic systems in the metal complex increases the catalytic activity towards

NADH oxidation. The best catalytic effect was observed for modification with the

complex B attached at the surface through the EDDA linker. This might be

rationalised by long chain of the EDDA linker, which provides better accessibility of

the mediator for the bulky NADH molecule and the presence of the 2,2’-bipyridine

ligand in the structure of the complex C might also affect the catalytic activity

towards NADH oxidation.

The values of inorm presented in Figure 3.9 were calculated based on surface

coverage of the phendione ligands, which is valid if we assume that all of the

phendiones at the surface are involved in the catalytic oxidation of the NADH. For

the octahedral complex C, the cis geometrical orientation of the two phendione

ligands might limit access of both quinone groups simultaneously to the relatively

large NADH molecule and prevent two molecules of the NADH being oxidised by

the pair of phendione ligands in a single C complex at the same time.


116

BAEDDA

HDAEDA

0

10

20

30

40

50

60

A

B

C

D

E i no

rm / 1

06 A

cm

5 mol- 2

linker

complex

b)

EDA HDA EDDA BA0

20

40

60

80

complex A

complex B

complex C

complex D

complex E

i no

rm / 1

06 A

cm

5 m

ol-2

linker

a)

Figure 3.9 Average values of the normalised catalytic currents inorm calcultated for 20 different

modifications in the library based on the surface coverage of the phendione ligands using

experimental data recorded for 1 mM NADH in phopshate buffer solution at scan rate of 50 mV s−1; a)

Each bar represents an average value of inorm and the standard errors of the mean calculated for 3

replicate electrodes; b) The means of inorm values presented in a three dimensional bar plot.

In order to present the experimental data in different way, the values of inorm were

recalculated based on the assumption that only one of the phendione ligands in the

complex C participates in the oxidation of NADH. In this case, the catalytic currents

were normalised by surface coverage of the metal complex as one molecule of the

complex C can catalyse oxidation of one molecule of the NADH at a time.

As shown in Figure 3.10, the values of inorm for the complex C are significantly

higher than for others metal complexes in the library. According to this assumption

the highest inorm of 72.5 and 82.4 × 106 A cm5 mol−2 were obtained for the complex C

attached at the surface through the EDA and EDDA linker, respectively. These

results clearly suggest that the presence of the additional phendione ligand

coordinated to the metal centre might contribute to the catalysis of NADH oxidation

at the GC surface.


117

BAEDDA

HDA

EDA0

18

36

54

72

A

B

C

D

E

i no

rm / 1

06 A

cm

5 mol-2

linker

complex

b)

EDA HDA EDDA BA0

11

22

33

44

55

66

77

88

99

in

orm

/ 1

06

Acm

5 m

ol-2

linker

complex A

complex B

complex C

complex D

complex E

a)

Figure 3.10 Average values of the normalised catalytic currents inorm calcultated for 20 different

modifications in the library based on the surface coverage of the metal complexes using experimental

data recorded for 1 mM NADH in 0.1 M phopshate buffer solution at scan rate of 50 mV s−1.

Complexes A, B, D and E contain one phendione ligand whereas complex C has two phendione

ligands; a) Each bar represents an average value of inorm and the standard errors of the mean calculated

for 3 replicate electrodes; b) The means of inorm values presented in a three dimensional bar plot.


118


119

3.5 Control electrodes.

Interesting results were observed for control electrodes 16-18 modified with

electrochemically inactive (2,2’-bipyridine)zinc (II) chloride complex and EDA

linker. Screening of these electrodes in 1 mM NADH solution gave faradaic current

at potential of 0 V, which corresponds to the oxidation of NADH catalysed by

quinone groups (Figure 3.11). In case of the control modifications, the only source of

quinone groups are those present at the oxidised glassy carbon surface, which are

known to catalyse the NADH oxidation at low overpotential.99

-0.3 -0.2 -0.1 0.0 0.1 0.2-10

-8

-6

-4

-2

0

2

4

6

i/ u

A

E vs. SCE/ V

electrode 10

-0.3 -0.2 -0.1 0.0 0.1 0.2-10

-8

-6

-4

-2

0

2

4

6

i/ u

A

E vs. SCE/ V

control electrode 16

-0.3 -0.2 -0.1 0.0 0.1 0.2-10

-8

-6

-4

-2

0

2

4

6

i/uA

E vs. SCE/ V


-0.3 -0.2 -0.1 0.0 0.1 0.2-10

-8

-6

-4

-2

0

2

4

6

i/ u

A

E vs. SCE/ V


Figure 3.11 Cyclic voltammograms for the modified electrode 13a and the control electrodes 70a-c

obtained duirng HTP screening in 0.1 M phosphate buffer solution pH 7, scan rate of 50 mV s−1 (black

curve), in 1 mM NADH aqueous solution at scan rate of 50 mV s−1 (red curve).

An average value of icat measured for the control electrodes is about 1.5 µA and it is

relatively low in comparison with the catalytic currents obtained for the

modifications with the EDA linker and zinc complex D with an average value of icat

about 3 µA. This indicates that the catalytic currents observed for all electrodes

modified with complexes from A to E are result of the catalysis of NADH by the

quinone groups of phendione ligand. As shown in Figure 3.11, the peaks for

oxidation of NADH by quinone at the surface and quinone groups of the phendione

occur at the same potential of 0 V. Therefore there is not possible to distinguish

between the catalysis come from quinone of metal complexes and the oxidised

carbon. In addition, the fraction of the NADH current catalysed by quinone directly

bonded at the surface might also depend on size of the linker and the metal complex.

As a result, values of icat and inorm for modification of electrodes by complexes A, B,

C, D and E discussed in Sections 2 and 3.4 also might include unknown amount of


120

the catalytic current obtained from oxidation of NADH by quinone functional groups

directly bonded at the surface.


121

3.6 Kinetic analysis for 63 member library

3.6.1 General

HTP electrochemical screening of the library of 63 modified electrodes in NADH

containing solution clearly showed that the complexes A-E bound at the electrode

surface catalyse oxidation of NADH at low overpotential (Sections 2 and 3). The

values of catalytic currents in the presence of NADH vary between different

modifications and clearly indicate effect of the linker and metal complex on the

catalytic activity towards NADH oxidation. In order to understand in more detail the

mechanism of the catalytic reaction between modified electrode surface and NADH

as well as justify the catalytic efficiency of individual modifications, the kinetics of

the NADH oxidation were studied for modified electrodes A-E.

In general, catalytic oxidation of NADH by mediator Mox can be described by a

reaction sequence obeying the well-known Michaelis-Menten kinetic model (Section

1.5). Reversible formation of an intermediate charge complex {NADH-Mox} allows

for a hydride transfer reaction within the complex followed by its decomposition to

yield reduced mediator Mred and NAD+ with a heterogeneous rate constant kcat.43

M cat +ox redNADH + Med {NADH-Mediator} NAD + Med

K k→ →←

Figure 3.12 Mechanism of NADH catalytic oxidation according to the Michaelis-Menten model.

This type of mechanism has been suggested for NADH oxidation at a number of

chemically modified electrodes. The presence of the intermediate complex between

mediators at the surface and NADH in solution was indicated for electrodes modified

by conducting organic salts,120 phenoxazine dyes,108,159,160 such as Meldola blue or

1,2-benzophenoxazine7-one (BPO) adsorbed at the electrode surface. The theoretical

kinetic model of the heterogeneous catalytic reaction was initially studied by Albery

and co-workers suggesting the analogy between the enzyme electrode kinetics and

heterogeneous redox catalysis.120,161-167 As a result, they developed a mathematical

model of the kinetics for heterogeneous catalysis where the amperometric response is

governed by number of the surface processes, such as:

• Mass transport of the substrate to the electrode surface including

concentration polarisation effect


122

• Bonding of the substrate to the mediator at surface to create an intermediate

reaction complex, if the catalytic reaction occurs according to Michaelis-

Menten model

• Electron and proton transfer within the intermediate complex followed by its

decomposition to yield products of the catalytic reaction

• Electrochemical regeneration of the mediator at the electrode surface

• Diffusion of the product from the surface considering the effect of product

inhibition.

Analytical treatment of the kinetics for heterogeneous catalysis proposed by Albery

was also applied for investigation of the kinetics of NADH and ascorbate oxidised at

polymer coated electrodes.7,9,10 In this case, the substrate is assumed to diffuse into

vacant sites within the polymer film, where the Michaelis-Menten type of reaction

occurs and the kinetics of the catalytic reaction were also determined by the

thickness of the film and the diffusion process of the substrate into the polymer.

Aspects of the kinetic studies of Albery were also applied by Lyons168 and Bartlett169

in a proposed theoretical model for oxidation of formate and glucose at hydrated

oxide electrodes in alkaline solution.168,170 The same theoretical model was applied

for kinetic analysis of the library of 63 modified electrodes presented in this thesis.

The following Section gives a brief explanation of the mathematical approach and

discussion of its application to twenty different modifications in the library.


123

3.6.2 Theoretical kinetic model for NADH oxidation at modified electrodes.168

The Michaelis-Menten kinetic approximation was used by Lyons168 and Bartlett169 in

order to design a kinetic model for heterogeneous redox catalysis for NADH

oxidation at stationary electrodes. According to their approach, the Michealis-

Menten catalytic reaction occurs at the solution/electrode interface, where NADH

diffuses to the electrode surface then forms the intermediate complex with the

immobilised mediator. This is followed by dissociation of the NAD+ and subsequent

electrochemical oxidation of the mediator to complete the catalytic cycle. The overall

scheme for the catalytic reaction is analogous to that discussed by Bartlett (Figure

3.13).169

[ ] [ ]

[ ] [ ]

[ ] [ ]

D

M

catO

DO

E

o

o

+

+ +

+

NADH NADH

NADH + Q NADH-Q

NADH-Q NAD + QH

NAD NAD

QH Q + 2e + H

k

K

k

k

k

∞

∞

→

→←

→

→

→

Figure 3.13 Proposed reaction scheme for the oxidation of NADH at the mediator modified

electrodes, according to Bartlett and Wallace.169 Subscripts ∞ and O indicate NAD in the bulk

solution and at the surface/solution interface. Q represents the oxidised form of the mediator;

QH represents the reduced form of the mediator; [NADH-Q] represents the intermediate

complex; kD is the diffusion rate constant and is considered to have the same values for

NADH and NAD+; KM is the Michaelis-Menten equlibrium constant for the complex

[NADH-Q]; kcat is heterogenous rate constant for the chemical reaction within [NADH-Q]

complex; kE is the rate constant for the electrochemical oxidation of the reduced mediator

QH.

As shown in Figure 3.13 NADH∞ diffuses from bulk solution to the surface/solution

interface NADHo with a diffusion rate constant kD. [Q] represents the oxidised

quinone groups of the phendione ligands, which binds the NADHo to give the

[NADH-Q] complex. The strength of the interactions between Q and NADH is

described by the equilibrium constant KM (units of mol cm−3). The bounded NADH

is oxidised by the quinone groups, giving the reaction products NAD+ and the

reduced form of mediator [QH]. Chemical reaction within the [NADH-Q] complex


124

occurs with a rate constant kcat (units of s−1). The reduced form of the mediator [QH]

is recycled electrochemically driven by the applied anodic potential at the electrode.

Electron transfer between the electrode and the immobilised mediator is assumed to

be rapid and its kinetics are described by the rate constant kE (units of s−1). Finally,

the NAD+ diffuses away from the surface with rate constant kD and product

inhibition by NAD+ in this case is ignored. The mass transport rate constants kD of

the substrate and product are assumed to be the same.

Mathematical analysis reported by Lyons leads to an expression for the steady-state

reaction rate or flux j, where the different possible rate-limiting steps can be

distinguished

M

D E med cat med cat med D

1 1 1 11

[NADH] [NADH] [NADH]

Kj

j k k k k k

= − + + +

Γ Γ Γ

3.1

where term [NADH] represents the concentration of NADH in a bulk solution and

Γmed is a surface coverage of the mediator. The overall flux j is determined by the

rate-limiting step in the catalytic reaction scheme, Equation 3.1 can be simplified for

each case. When mass transport of NADH is slow, the term in kD will be dominant.

The overall flux can also be limited by the kinetics of the chemical reaction between

mediator and NADH. For lower concentrations of NADH, the flux is limited by the

binding constant between product and mediator and the catalytic rate constant given

by KM and kcat, respectively. Equations can be then reduced to

M

cat med

1[NADH]

K

j k=

Γ

3.2

In the case of saturated kinetics at high NADH concentrations, all quinone groups at

the surface can take part in the catalytic reaction and the expression for j will be

reduced to

cat med

1 1j k

=Γ

3.3

When the overall reaction rate is governed by the rate of the electron transfer

between the electrode and mediator kE, then the Equation 3.1 reduces to

E med

1 1j k

=Γ

3.4


125

In the case of the slow NADH transport through the diffusion layer, concentration

polarisation will cause the surface concentration of NADH to be less than the

concentration of NADH in the bulk solution and results in less saturated catalysis

than would be expected. This effect is described in Equation 3.1 as

D

11

[NADH]j

j k= −

3.5

Next, using the relationship between current and flux,171

nFAi j= 3.6

Equation 3.1 was rearranged by Lyons and Bartlett168 to give an expression for the

overall catalytic current icat

( ) ( )2D ME ME

cat ME MED

4 [NADH ]F + [NADH ] + [NADH ]

2

k K ki n A K K

k

∞∞ ∞

= − −

3.7

where KME is the Michaelis-Menten constant for the modified electrode defined by

1

MME

cat D E cat

1 1 =

KK

k k k k

− Γ

+ +

3.8

and kME is the effective electrochemical rate constant for modified electrode at low

substrate concentrations

1

MME

cat D

1 =

Kk

k k

−

+ Γ

3.9

The above expression for the catalytic current has four limiting cases depending on

the relative values of NADH concentration and the different kinetic parameters.

I. Mass transport kinetics

When the mass transport of NADH to the electrode is the limiting factor, the rate

constant kD will be dominant in the expression for the overall current. This is valid

only for unsaturated catalysis ([NADH]<KME) when the catalytic reaction and the

electrochemical regeneration of the mediator are fast enough not to be considered as

rate limiting (kD<kcatΓ/KM). In this case the current can be described

I D bulknFA [NADH]i k=

3.10


126

In this work, the diffusion rate constant kD was calculated using the Randles-Sevčik

equation for the mass transport limited current at a stationary electrodes172

D 0.46 n F

k v DRT

=

3.11

F, R and T have their usual meanings, v is the scan rate (V s−1), n is the number of

electrons taking part in the electrochemical reaction and D represents diffusion

coefficient, which in this case equals 2.4×10−6 cm2 s−1 for NADH in aqueous

solution.173

II. Unsaturated reaction kinetics kcat/KM

In this case, the overall current is determined by the term in kcat/KM

cat bulkII

M

nFA [NADH]ki

K

Γ=

3.12

In this situation, the kinetics of the chemical reaction are slower than the mass

transport of the NADH (kD>kcat/KM) and electrochemical reaction (kE>kcat/KM).

III. Saturated catalysis and slow electrochemical reaction kinetics

For higher concentrations of NADH, where all the active sites of the mediators take

part in catalysis ([NADH]>KME), if the rate of mass transport and the catalytic rate

constant are assumed to be greater than the rate of electron transfer between mediator

and surface (kE<kcat and kE<kD). The overall current will be described by

III Ei nFAk= Γ

3.13

IV. Saturated catalysis and slow chemical reaction kinetics

In the case of saturated catalysis at high NADH concentrations, the kinetics of the

chemical reaction can also become the limiting step if the electrochemical and mass

transport kinetics are fast (kE>kcat and kE>kD). In this case, the chemical reaction within

the intermediate complex between NADH and quinone groups is rate limiting. The

current in this situation is given by

IV catnFAi k= Γ 3.14


127

3.6.3 Analysis of the heterogeneous rate constant kcat and the equilibrium

constant KM for the library of 63 modified electrodes.

The mathematical model presented in Section 3.6.2 was applied to analyse the

experimental data obtained from the HTP screening of the library, carried out at four

different scan rates 10, 20 50 and 100 mV s−1 and a range of NADH concentrations

from 1 to 4 mM. Plots of the catalytic currents as function of NADH concentration

were fitted to the Equation 3.7, using commercial software (Origin 8.1). Figure 3.14

shows a typical example of the fitting curves for electrodes 13a-c.

0 1 2 3 4-1

0

1

2

3

4

5

6

7

13a electrode

13b electrode

13c electrode

fit 13a electrode

fit 13b electrode

fit 13c electrode

i ca

t / µ

A

NADH /mM

Figure 3.14 Plot of catalytic current icat as a function of NADH concentration at a scan rate 10 mV s−1

obtained for modifed electrodes 13a-c; the points represent the experimental data and the lines are the

best fits calculated from the theoretical model using Equation3.7.

Fitting of the theoretical model was also successfully obtained for experimental data

at different scan rates, see Figure 3.15a for an example. However, the catalytic

current at 100 mV s−1 in 2 mM NADH solution gives unexpected the lower values

and leads to poor quality of the fitting curve. A similar problem was observed for

most of the library members, which might be related to some unknown instrumental

error during measurement. Fitting of the 100 mV s−1 curve was improved after

excluding the data for 2 mM NADH solution for all members of the library (Figure

3.15b).

A similar problem occurred for some of the other experimental data, mostly for the

electrodes modified with the BA linker at slow scan rates. The source of these errors

is unknown. To obtain the best fits, these problematic data were excluded from our


128

calculations. Values of kME and KME for the best fits calculated for all members of the

library are given in Appendixes 4-7.

0 1 2 3 4

0

2

4

6

8

10

10 mV/s

20 mV/s

50 mV/s

100 mV/s

fit 10 mV/s

fit 20 mV/s

fit 50 mV/s

fit 100 mV/s

i cat /

µA

NADH /mM

a)

0 1 2 3 4

0

2

4

6

8

10

experimental data with 2 mM NADH

experimental data without 2 mM NADH

fitting curve without 2 mM NADH

fitting curve with 2 mM NADH

i ca

t / µ

A

NADH /mM

b)

Figure 3.15 a) Plot of catalytic current as a function of NADH concentration at scan rates of 10, 20,

50 and 100 mV s−1; the points are experimental data, the lines are fitting curves obtained calculated

from Equation 3.7; b) Comparison of the fitting curves obtained for scan rate 100 mV s−1 with

experimental data in 2 mM NADH (black line) and without 2 mM NADH (red line).

Rearrangement of Equation 3.8 and 3.9 leads to expressions for the catalytic rate

constant kcat and the Michaelis-Menten equilibrium constant KM based on the

assumption that the electrochemical regeneration of mediator at the electrode surface

is not the rate-limiting step (kE > kD and kE > kcat). The rate of electron transfer

between electrode surface and mediator is relatively rapid as evidenced by constant

peak potentials observed for all members of the library in range of applied scan rates.

For values of kE → ∞, Equation 3.8

1

MME

cat D cat

1 =

KK

k k k

− Γ

+

3.15

which can be rewritten as

medM ME

cat cat D

Γ=

K K

k k k−

3.16

Equation 3.16 was then substituted into Equation 3.9 to obtain an expression for the

catalytic rate constant kcat

1

' medMEME

cat D med D

1 1Kk

k k k

− Γ

= − + Γ

3.17


129

'ME ME

catmed

k Kk =

Γ

3.18

Rearrangement of Equations 3.15 and 3.18 gave the expression for the Michaelis-

Menten constant

MEM ME

D

1k

K Kk

= −

3.19

Values of the fitting parameters kME, KME and the calculated values of kcat and

KM for all the electrodes in the library are presented in Appendix 8. Each value of

kME and KME was calculated using experimental data obtained for three different

NADH concentrations (1 mM, 2 mM and 4 mM) and screened at four different scan

rates (10 mV s−1, 20 mV s−1, 50 mV s−1 and 100 mV s−1). The values of kME and KME

for each modified electrode in the library were then used to calculate kcat and KM.

Average values of kcat and KM for twenty different modifications were calculated

using three replicate electrodes and their summaries are presented in Figure 3.16-

3.18. Overall, the mean value of kcat or KM for each modification represents

experimental data obtained from 36 single sets of HTP measurments. The HTP

electrochemical screening allowed for rapid evaluation and direct comparison of the

kinetics between different modifications.

In case of the kcat, structure of the metal complex has a pronounced effect on

the heterogeneous rate constants kcat (Figure 3.16). The highest values of kcat were

obtained for modifications with ruthenium complexes B and C with additional

bidentate ligands such as 2,2’-bipyridine or phendione. In comparison, the values of

kcat obtained for electrodes modified with complex A containing two monodentate

chloride ligands are relatively lower. This might indicate that the presence of the

bidentate ligand with the conjugated aromatic system increases the rate of the

catalytic reaction in the bound complex.

As seen from Figure 3.16, choice of the linker also affects the rates of the catalytic

reaction. The highest calculated value of kcat was obtained for modifications with

relatively long EDDA, containing ethoxy functional groups in its structure.

Relatively low values of kcat were observed for modifications with the EDA linker.

These results would clearly suggest that the longer linker increases the rate of the

catalytic reaction, which might be rationalised by better accessibility of the bulk


130

metal complex to the large molecule of NADH. However, the proposed

rationalisation could not be applied for modifications with the longer aliphatic linker

HDA, where the values of kcat were lower in comparison to the modifications with

shorter EDA linker. For a given metal complex, the lowest values of kcat were

observed for the modifications with BA linker as would be expected from very low

values of icat recorded for these modifications.

Figure 3.16 Average values of catalytic rate constants kcat caclulated using surface coverage of the

phendione ligands obtained for 20 different modifications in the library, where kcat for each individual

electrode was calculated using experimental data of 12 single sets of HTP measurments performed in

0.1 M phosphate buffer solution pH 7; complexes A, B, D and E contain one phendione ligand

whereas complex C has two phendione ligands; a) means values of kcat with the standard error of the

mean of three replicate electro`des; b) 3D bar plot of average values of kcat for 20 different

modifications in the library.

According to Equation 3.18, the catalytic rate constants were normalised by

surface coverage of the phendione ligands. For complex C, these results are based on

the assumption that both of the phendione ligands participate in the catalytic

oxidation of two molecules of NADH. However, oxidation of two molecules of

NADH by both of the phendione ligands might be limited due to the cis-geometry of

this complex, which might result in lower accessibility of the bulky NADH

molecules to the quinone functional groups. Therefore, a second assumption was

suggested in the analysis of kcat, where only one phendione ligand in the metal

complex participates in the catalytic oxidation of NADH. In this case, the catalytic

BAEDDA

HDAEDA

0

4

8

12

16

A

B

C

D

E

k cat / s

- 1

linkercomplex

b)

EDA HDA EDDA BA0

5

10

15

20

com plex A

com plex B

com plex C

com plex D

com plex E

kca

t / s

-1

linker

a)


131

rate constants are normalised by surface coverage of the metal center as the whole

molecule of the complex C participates in oxidation of molecule of the NADH. As a

result, the calculated values of kcat for the complex C increase their values by a factor

of two making it by far the most efficient mediator towards NADH oxidation (Figure

3.17).

EDA HDA EDDA BA0.0

2.8

5.6

8.4

11.2

14.0

16.8

19.6

22.4 complex A

complex B

complex C

complex D

complex E

kca

t / s

-1

linker

a)

Figure 3.17 Average values of catalytic rate constants kcat caclulated using surface coverage of the

metal complex obtained for 20 different modifications in the library. The value of kcat for each

individual electrode was calculated using experimental data for 12 single sets of HTP measurments

performed in 0.1 M phosphate buffer solution pH 7; a) mean values of kcat with the standard error of

the mean for the three replicate electrodes; b) 3D bar plot of average values of kcat for 20 different

modifications in the library.

The equilibrium constants KM represent the strength of the intermediate

complex formed between the NADH and the quinone groups. In general, values of

the equilibrium constant KM calculated for all modifications in the library are

consistent for the electrodes modified by the aliphatic linkers EDA, HDA and EDDA

at approximate KM of 1.8 mM (Figure 3.18). A significant decrease in KM was

observed for modifications with the BA linker, which indicates that the charge

transfer complex is more stable for the BA linker than for the aliphatic linkers. In this

case, the relatively small amount of metal complex attached through this linker at the

electrode surface might increase the number of uncoordinated 2,2’-bipyridine ligands

present at the surface. This might additionally stabilises the intermediate complex by

BAEDDA

HDAEDA

0

5

10

15

20

A

B

C

D

E

kcat /s

-1

linker

complex

b)


132

interactions between free 2,2’-bipyridine ligands and the phosphate groups of the

NADH.

EDA HDA EDDA BA0.0

0.5

1.0

1.5

2.0

2.5

complex A

complex B

complex C

complex D

complex E

KM / m

M

linker

a)

Figure 3.18 Average values of Michaelis-Menten constnat KM and their three dimensional

representation obtained for 20 different modifications in the library. Each bar represents mean value

of KM and standard error of the mean of three replicate electrodes, where KM for each individual

electrode was calculated using experimental data of 12 single sets of HTP measurments performed in

0.1 M phosphate buffer solution pH 7.

To the best of our knowledge, the discussed values of kcat and KM are the first

report of quantitative analysis of the NADH catalysis by the phendione containing

metal complexes based on the theoretical kinetic model proposed by Bartlett and

Lyons. This theoretical kinetic model allowed us to evaluate the Michaelis-Menten

constants and the first order rate constants kcat corresponding to the saturated

catalysis where the kinetics are independent on the NADH concentration. A similar

kinetic model was also developed for oxidation of NADH at polymer coated

electrodes, where the catalytic reaction occurs within the whole polymer film rather

than at the polymer surface and the reaction is reversibly inhibited by the oxidation

product NAD+.169,174 For the reported examples of polymer mediators, the values of

rate constants were determined as a function of film thickness, electrode potential

and NAD+ concentration. Therefore, the literature examples could not be directly

compared with the kinetic parameters obtained for the phendione complexes

presented in the library, where the values of kcat were determined based on

assumptions that the reaction occurs at the electrode/solution interface and the NAD+

inhibition was neglected.

BAEDDA

HDAEDA

0.0

0.5

1.0

1.5

2.0

2.5

A

B

C

D

E

KM

/m

ol cm

-3

linker

complex

b)


133

However, a number of the phendione containing metal complexes physically

adsorbed at the electrode surface were reported in the literature and showed

significant amperometric response for NADH oxidation.116,126,127,131,132,134,175 Kinetic

studies for these complexes were based on a simplified kinetic model, where mass

transport of the NADH is fast and concentration polarisation is negligible. The

kinetic analysis reported for these complexes describes second order catalytic rate

constants valid for low concentrations of NADH ([NADH] < KM), where the

chemical reaction between NADH and mediator is considered as bimolecular.

In order to compare the catalytic efficiency with the literature examples, the values

of kcat normalised by surface coverage of the metal complexes and the values of KM

were used to calculate a rate constant kcat/KM corresponding to unsaturated catalysis

at low NADH concentrations ([NADH] < KM). In this case, the reaction between

quinone groups and the NADH is considered as bimolecular and the calculated rate

constant kcat/KM is assumed to be second order (units of M−1 s−1).

Figure 3.19a clearly shows that the values of kcat/KM depend on the metal complex

for electrodes modified with the aliphatic linkers. Length and structure of the

aliphatic linker has a minor effect on the values of kcat/KM. The best catalytic

efficiency was obtained for electrodes modified with the complex C and EDDA

linker with the highest kcat/KM of 7.44×103 M−1 s−1. In general, high values of kcat/KM

were obtained for complexes B and C in comparison with the other metal complexes.

The effect of the linker for reaction rate at low NADH concentrations is negligible.

In addition, relatively high average values of kcat/KM were observed for modifications

with the BA linker, however high errors obtained for this linker unable compare

them with other modifications in the library. For complex C, the catalytic rate

constants kcat were also calculated based on the assumption that only one phendione

ligand participates in the catalytic reaction giving values of kcat increased by a factor

of two (Figure 3.16).


134

EDA HDA EDDA BA0

2

4

6

8

10

12

14

16

complex A

complex B

complex C

complex D

complex E

kca

t / K

M / 1

03 M

-1 s

-1

linker

a)

Figure 3.19 Average values of rate constants kcat/KM at low NADH concentrations ([NADH] < KM)

obtained for 20 different modifications in the library; Each bar represents mean values of kcat/KM with

a standard error of the mean for three replicate electrodes; a) Means for kcat/KM calculated using values

of kcat presented in Figure 3.16 for the surface coverage of phendione ligands and KM in Figure 3.18;

b) Means for kcat/KM calculated using values of kcat presented in Figure 3.17 for the surface coverage

of metal complexes and KM in Figure 3.18.

The values of kcat/KM for the library members were compared with a number

of literature examples for modified electrodes with adsorbed monolayers of

phendione containing metal complexes. The reported rate constants for unsaturated

catalysis suggest that metal complexes with a higher number of phendione ligands

increase catalytic efficiency and vary from 0.9×103 (M−1 s−1) for [Os(4,4’-dimethyl-

2,2’-bipyridine)2(phendione)]2+ and 1.2×103 (M−1 s−1) for [Ru(phendione)2(5-amino-

1,10-phenanthroline)]2+ to 6.8×103 (M−1 s−1) for a monolayer of [Fe(phendione)3]2+

complex.134 The same effect was observed between modifications in the library,

where complexes A, B, D and E with one phendione ligand gave lower values of

kcat/KM than electrodes modified with complex C containing two phendione ligands

(Figure 3.19). In comparison to the literature examples, the values of kcat/KM obtained

for the complexes A, B, D and E attached at the surface through different linkers are

relatively high. Similarly, second order rate constants for the modifications with

complex C are significantly higher than kcat/KM reported for the literature examples.

The pronounced catalytic efficiency observed for complex C in the case of only one

phendione participating in the catalytic reaction (Figure 3.19b) is about two times

higher than the catalytic efficiency reported for the [Fe(phendione)3]2+ complex.

EDA HDA EDDA BA0

2

4

6

8

10

12

14

16 complex A

complex B

complex C

complex D

complex E

kc

at /

KM / 1

03 M

-1 s

-1

linker

b)


135

These results would clearly suggest that the molecular architecture of the surface by

combination of the metal complex and linker have a pronounced effect on the

catalytic efficiency kcat/KM towards catalytic oxidation of NADH.


136


137

3.7 Conclusion

The library of 63 modified electrodes was designed based on previous results

obtained for the individual electrodes and successfully prepared using combinatorial

and solid-phase synthesis methods. To the best of our knowledge, the presented

library of covalently modified electrodes is the first report on molecular design of the

phendione type of mediators covalently bound at the electrode surface in a

combinatorial way. HTP electrochemical screening allowed for evaluation and direct

comparison of electrochemical properties and catalytic activity for all modified

electrodes in the library in a single set of measurements.

Firstly, the library was screened in aqueous solution to evaluate the amount of the

metal complex bound at the surface. Comparison of the calculated surface coverage

between different modifications clearly suggested that the structure of the metal

complex and type of the linker have an impact on the amount of the bound complex

at the surface. The coverage decreased with the size and geometry of the metal

complex with relatively high values of surface coverage obtained for tetrahedral zinc

complexes D and E and low coverages were found for modifications with bulky

octahedral ruthenium tris-complexes B and C. For modifications with aliphatic

linkers, the coverage of metal complexes tends to decrease with longer aliphatic

chain. Relatively small amounts of the bound complex were obtained for the rigid

benzylamine linker BA, where limited flexibility of this linker decreases the amount

of the complex bound at the surface.

HTP screening of the library in the presence of NADH allowed for direct comparison

of the catalytic activity between different modifications. In general, the values of

directly measured catalytic currents icat are proportional to the surface coverage,

where high catalytic currents were recorded for modification with relatively small

tetrahedral zinc complexes D and E and lower values of catalytic currents were

observed for octahedral ruthenium complexes A, B and C.

Catalytic currents normalised by surface coverage of the metal complex or phendione

ligands clearly showed that the presence of tris-ruthenium complex containing

ligands with conjugated aromatic systems gave pronounced catalytic activity toward

NADH oxidation. For coverage of phendione ligands, the highest catalytic activity

was observed for modification with complex B and EDDA. For the coverage of


138

metal complex, the highest catalytic activity was observed for complex C attached at

the surface through EDDA linker.

Similar results were obtained from studies of the kinetics for catalytic reaction in the

presence of NADH based on a theoretical model for heterogeneous catalysis. The

HTP electrochemical screening allowed for rapid evaluations of the heterogeneous

rate constants using experimental data obtained for three different NADH

concentrations at several scan rates. For rate constants in the case of saturated

catalysis, comparison between different modifications in the library clearly suggested

that the reaction rates depend on the geometry and structure of the metal complex.

Any effect of the linker on the reaction rate is negligible. Relatively high rate

constants were calculated for octahedral ruthenium complexes B and C with an

additional bidentate ligand coordinated to the metal ion. Equally high values of the

rate constants were obtained for modifications with the ruthenium complex C

attached at the surface through EDA and EDDA linker.

In addition, catalytic rate constants obtained for twenty different modifications in the

library in the case of unsaturated catalysis were significantly higher than reported

literature examples of electrocatalysis of NADH oxidation by the metal complexes

physically adsorbed at electrode surface.

In conclusion, the presented catalytic analysis for different modifications in the

library suggests that molecular design of the surface electrode allows for control over

the catalytic activity and kinetics of the catalytic reaction between mediator at the

surface and NADH. The array of mediator systems analysed in the library, clearly

suggests that the modifications with the highest catalytic activity are attractive

mediators in biosensor applications. All of the modified electrodes regenerate the

NADH to enzymatically active NAD+ at low overpotentials, however they exhibit

differences in the catalytic response. The electrode modified with complex E

attached at the surface through the EDA linker generates the highest catalytic current

corresponding to the number of NADH molecules being regenerated to NAD+ during

the catalytic cycle. Therefore this mediator would be the best choice to apply for

activity-based biosensors.

However, the most effective mediator was found to be modifications with complex B

or C and the EDDA linker, since these modifications gave the highest catalytic


139

activity independent on the coverage of mediator (inorm) and fastest kinetics during

regeneration of the NADH to enzymatically active NAD+.

To the best of our knowledge, the presented analytical approach is the first

report on the evaluation of library of mediators for NADH catalytic oxidation using

HTP electrochemical screening. The HTP screening allowed for rapid evaluation of

the electrochemical and catalytic properties of all modified electrodes and selection

of modified electrodes with the best catalytic properties towards NADH oxidation,

which could be used in the biosensor applications.


140

Experimental Section

141

4 Experimental Section

4.1 Synthesis

4.1.1 General

Synthetic experiments were carried out under air unless otherwise specified.

Experiments requiring dry conditions were performed under nitrogen atmosphere in

dried glassware. Solvents and reagents were commercial grade and they were used

without further purification or where necessary were distilled prior to use. DMF was

distilled under reduced pressure and stored in a Schlenk bottle over M/S 4Å. Ethanol

and DCM were distilled above calcium hydride directly before use. Dry THF was

obtained by distillation of THF and benzophenone mixture above sodium wire under

the argon atmosphere.

Flash chromatography columns were performed on Silica 60A, particle size 35-70

micron (Fisher Scientific). Thin layer chromatography was performed on silica pre-

coated aluminium plates (Merck Silica gel F254) and the spots were visualised with

UV light or KMnO4 stain.

The following compounds were synthesised according to the literature procedures:

linkers 4-(N-Boc-aminomethyl)benzene diazonium tetrabluoroborate salt147,148 and

tert-butyl N-(aminomethyl)phenylmethyl)carbamate,176 ligands 1,10-phenanthroline-

5,6-dione177 and 2,2’-bipyridine-5-carboxylic acid,143 complexes

dichlorotetrakis(dimethyl sulphoxide) ruthenium (II)178 and (1,10-phenanthroline-

5,6-dione)zinc (II) chloride.150

4.1.2 Instrumentation

Proton NMR were obtained at 300 MHz on a Bruker AC 300 and 400 MHz on a

Bruker DPX 400 spectrometer. Carbon NMR spectra were recorded at 75.5 MHz on

Bruker AC 300 and 100 MHz on Bruker DPX 400. Chemical shifts (δ) are reported

in ppm and coupling constants (J) are given in Hz. Spectra were referenced with

respect to the residual peak for the deuterated solvent. The following abbreviations

are used for spin multiplicity: s = singlet, d = doublet, t = triplet, q = quadruplet, m =


142

multiplet and br = broad signal. For metal complexes, assignment abbreviations are:

pd for 1,10-phenanthroline-5,6-dione ligand and bpy for 2,2’-bipyridine ligand.

Carbon spectra were proton decoupled and the multiplicities of the signals quoted

within the brackets using the following notation: primary carbon (3), secondary (2),

tertiary (1) and quaternary (0).

Infra-red spectra were recorded on BIORAD Golden Gate FTS 135. Melting points

were determined in open capillary tubes using a Gallenkamp Electrothermal melting

point apparatus. Mass spectra were obtained on VG analytical 70-250 SE normal

geometry double focusing mass spectrometer. All electrospray (ES) spectra were

recorded on a Micromass Platform quadruple mass analyser with an electrospray ion

source using acetonitrile as a solvent. High resolution accurate mass measurements

were carried out at 10,000 resolution on a Bruker Apex III FT-ICR mass

spectrometer. MALDI mass spectra were recorded at Micromass TofSpec2E

spectrometer.


143

4.1.3 Synthesis of 2-(2-(pyridin-2-yl)-1H-imidazol-1-yl)acetic acid ligand.

4.1.3.1 tert-Butyl 2-(2-(pyridin-2-yl)-1H-imidazol-1-yl)acetate.154

2-(1H-Imidazol-2-yl)pyridine (1.45 g, 10 mmol) was prepared according to literature

procedure157 and dissolved in dry THF (20 mL) under nitrogen. The solution was

cooled to −78 °C and n-butyl lithium (2.5 M in hexane) (4.5 mL, 11 mmol) was

added dropwise. The stirred reaction mixture was kept at −78 °C for 3 h and a

solution of tert-butylbromoacetate (1.6 mL, 11 mmol) in dry THF (20 mL) was then

added dropwise. The reaction mixture was allowed to warm to room temperature

followed by stirring overnight under nitrogen. The reaction mixture was quenched

with 10 mL of water and three times extracted with DCM (3 × 25 mL) The organic

layers were dried over magnesium sulphate and solvent was evaporated in vacuo The

residue was purified by flash chromatography (eluent MeOH : DCM, 1 : 9) to give

tert-butyl 2-(2-(pyridin-2-yl)-1H-imidazol-1-yl)acetate as a brown oil in 76 % yield

(2 g). Rf = 0.8 (eluent MeOH : DCM, 1 : 9), 1H NMR (300 MHz, CDCl3): δ = 8.49

(1 H, d, J = 4.8, PyH), 8.25 (1 H, d, J = 8.1, PyH), 7.74 (1 H, td, J = 7.7, 1.8, PyH),

7.18-7.22 (1 H, m, PyH), 7.17 (1 H, s, ImH), 6.97 (1 H, s, ImH), 5.20 (2 H, s, CH2),

1.39 (9 H, s, C(CH3)3), 13

C NMR (75 MHz, CDCl3): δ = 167.4 (0), 150.2 (0), 147.9

(1), 144.5 (0), 136.6 (1), 128.1 (1), 124.3 (1), 122.4 (1), 122.1 (1), 82.0 (0), 51.3 (2),

27.9 (3), IR (cm-1

): 2978 (v), 1743 (s), 1589 (m), 1152 (s), 1093 (m), 740 (m), 705

(m), LR-MS ES+ (m/z): 282.2 [M + Na]+ (100%), 260.2 [M + H]+ (47%), HR-MS

ES+ (m/z): found 260.1395 [M + H]+, calculated 260.1394.


144

4.1.3.2 2-(1-(carboxymethyl)-1H-imidazol-3-ium-2-yl)pyridin-1-ium

trifluoroacetate salt.

tert-Butyl 2-(2-(pyridin-2-yl)-1H-imidazol-1-yl)acetate (0.26 g, 1 mmol) was

dissolved in 20 % solution of TFA in DCM (15 mL) and stirred overnight at room

temperature under nitrogen atmosphere. The solvent was evaporated to give an oily

residue and azeotropic distillation in toluene and subsequent washing with DCM and

ethyl acetate gave 2-(1-(carboxymethyl)-1H-imidazol-3-ium-2-yl)pyridin-1-ium

trifluoroacetate salt as a bright yellow solid in 97 % yield (0.42 g). M.p. (crystallised

from MeOH) 158-162 °C. 1H NMR (300 MHz, DMSO-d6): δ = 12.2 (1 H, br. s, CO2H), 8.63-8.75 (1 H, m,

PyH), 8.13-8.21 (1 H, m, PyH), 8.17 (1 H, d, J = 8.1, PyH) 8.07 (1 H, td, J = 7.7,

1.8, PyH), 7.76 (1 H, d, J = 1.5, ImH), 7.66 (1 H, d, J = 1.5, ImH), 7.56 (1 H, dd, J =

7.7, 4.8, PyH), 5.45 (2 H, s, CH2), 13

C NMR (75 MHz, DMSO-d6): δ = 168.8 (0),

149.4 (1), 144.9 (0), 141.8 (0), 138.0 (1), 126.2 (1), 125.2 (1), 123.2 (1), 121.9 (1),

50.8 (2), IR (cm-1

): 3131 (v), 3165 (v), 1688 (m), 1582 (v), 1141 (s), 777 (s), 714 (s),

LR-MS ES+ (m/z): 204.2 [M – H]+ (100%), HR-MS ES+ (m/z): found 204.0770

[M– H]+, calculated 204.0768.

4.1.4 Synthesis of ruthenium (II) complexes.

4.1.4.1 Synthesis of bis-(dimethyl sulphoxide)(1,10-phenanthroline-5,6-dione)

ruthenium (II) chloride.179

Complex dichlorotetrakis(dimethyl sulphoxide) ruthenium (II) (0.1 g, 0.2 mmol) and

1,10-phenanthroline-5,6-dione ( 42 mg, 0.2 mmol) were dissolved in distilled ethanol


145

(3 mL) and refluxed overnight under nitrogen. The brown reaction mixture was

cooled to room temperature and precipitated brown solid was washed with diethyl

ether and dried under vacuum to give bis-(dimethyl sulphoxide)(1,10-

phenanthroline-5,6-dione) ruthenium (II) chloride in 55 % yield (0.06 g). M.p. above

300 °C. 1H NMR (400 MHz, DMSO-d6): δ = 9.84 (1 H, d, J = 5.5, CqCH-ring B), 9.72 (1 H,

d, J = 5.0, CqCH-ring A), 8.63 (1 H, d, J = 8.0, NCH-ring B), 8.49 (1 H, d, J = 8.0,

NCH-ring A), 7.98 (1 H, dd, J = 8.0, 5.5, (CH)2CH-ring B), 7.83 (1 H, dd, J = 8.0,

5.0, (CH)2CH-ring A), 3.43 (6 H, s, SO(CH3)2), 2.89 (3 H, s, SO(CH3)2), 2.33 (3 H,

s, SO(CH3)2), 13

C NMR (100 MHz, DMSO-d6): δ = 181.2 (0), 174.5 (0), 159.0 (1),

157.5 (0), 155.5 (1), 155.4 (1), 136.3 (1), 135.3 (1), 131.6 (0), 130.9 (0), 127.4 (1),

127.2 (1), 46.2 (3), 45.6 (3), 44.6 (3), 44.1 (3), IR (cm-1

):3085 (w), 3006 (w), 2920

(w), 1705 (m), 1431 (m), 1078 (s), 727 (m), 440 (w), MS-MALDI TOF (LD+)

(m/z): found 539.9 for [M]+, calculated 539.9.

4.1.5 Synthesis of (2,2’-bipyridine)(1,10-phenanthroline-5,6-dione) ruthenium

(II) chloride.142

Complex bis-(dimethyl sulphoxide)(1,10-phenanthroline-5,6-dione) ruthenium (II)

chloride (0.1 g, 0.2 mmol) and 2, 2’-bipyridine (31 mg, 0.2 mmol) were dissolved in

DMF (16 mL) and dark solution was heated under reflux for 4 h under nitrogen

atmosphere and reduced light. The dark purple solution was then evaporated to

dryness in vacuo and the residue dissolved in 4 mL of methanol. The purple solution

was left overnight at low temperature to give black-purple microcrystalline solid,

which was collected by filtration, washed ten times with 50 mL of diethyl ether and

dried in vacuo to give (2,2’-bipyridine)(1,10-phenanthroline-5,6-dione) ruthenium

(II) chloride as a microcrystalline solid in 60 % yield (60 mg). M.p. decomposed

above 300 °C. 1H NMR (400 MHz, DMSO-d6): δ = 10.08 (1H, d, J = 5.0, CqCH-


146

ring B pd), 10.01 (1 H, d, J = 5.5, CqCH-ring A pd), 8.66 (1 H, d, J = 8.5, CqCH-ring

B bpy), 8.51 (1 H, d, J = 8.0, CqCH-ring A bpy), 8.44 (1 H, J = 7.5, NCH-ring B pd),

8.11 (1 H, t, J = 8.3, (CH)2CH-ring B bpy),8.06 (1 H, d, J = 8.0, NCH-ring B bpy)

7.97 (1 H, dd, J = 7.8, 5.5, (CH)2CH-ring B pd), 7.79–7.85 (2 H, m, (CH)2CH and

NCH-ring A bpy), 7.73 (1 H, t, J = 7.5 (CH)2CH-ring B bpy), 7.51 (1 H, d, J = 5.5

Hz, NCH-ring A pd), 7.30 (1 H, dd, J = 7.5, 6.0, (CH)2CH-ring A pd), 7.14 (1 H, t, J

= 6.5, (CH)2CH-ring A bpy), 13C NMR (100 MHz, DMSO-d6): δ = 180.9 (0), 175.0

(0), 174.8 (0), 159.8 (0), 159.4 (0), 158.1 (0), 157.5 (0), 156.7 (1), 155.6 (1), 153.2

(1), 152.4 (1), 135.1 (1), 134.0 (1), 132.0 (1), 2 × 130.73 (1), 130.70 (1), 130.4 (0),

126.2 (1), 125.5 (1), 125.2 (1), 122.8 (1), 122.6 (1), IR (cm-1

): 2990 (v), 1688 (s),

1459 (v), 761 (s), 722 (s), MS-MALDI TOF (LD+) (m/z): found 503.3 [M – Cl]+,

calculated 503.0.

4.1.6 Synthesis of bis-(1,10-phenanthroline-5,6-dione)ruthenium (II)

chloride.142

Complex dichlorotetrakis(dimethyl sulphoxide) ruthenium (II) (0.1 g, 0.2 mmol) and

1,10-phenanthroline-5,6-dione (0.087 g, 0.4 mmol) were dissolved in dry DMF (25

mL) under nitrogen and heated under reflux for 4 hours. The black reaction mixture

was evaporated to dryness in vacuo and the black residue was dissolved in 4 mL of

methanol. The solution was left overnight at 5 °C to precipitate the black

microcrystalline solid, which was collected by filtration and washed with 50 mL of

diethyl ether followed by drying in vacuo to give bis-(1,10-phenanthroline-5,6-

dione)ruthenium (II) chloride in 33 % yield (0.04 g). M.p. above 300 °C. 1H NMR

(400 MHz, DMSO-d6): δ = 10.12 (2 H, d, J = 5.5, CqCH-ring B), 8.49 (2 H, d, J =

8.0, CqCH-ring A), 8.10 (2 H, d, J = 8.0, NCH-ring B), 8.02 (2 H, dd, J = 8.0, 5.5,

(CH)2CH-ring B) 7.77 (2 H, d, J = 5.5, NCH-ring A), 7.35 (2 H, dd, J = 8.0, 5.5,

(CH)2CH-ring A), 13C NMR (100 MHz, DMSO-d6): δ = 175.3 (0), 175.2 (0), 159.6


147

(0), 157.9 (0), 157.2 (1), 156.5 (1), 133.0 (1), 131.9 (1), 131.1 (0), 131.0 (0), 126.9

(1), 126.6 (1), IR (cm-1

): 1689 (s), 1557 (v), 1297 (m), 1103 (m), 833 (m), 824 (m),

MS-MALDI TOF (LD+) (m/z): found 557.3 [M – Cl-]+, calculated 557.0 [M – Cl- +

MeCN]+.


148

4.2 Electrochemical and solid-phase covalent modification of GC

electrodes.

4.2.1 Instrumentation

Electrochemical experiments for single electrodes were recorded on an Autolab

PGSTAT30 Potentiostat/Galvanostat (Eco Chemie, Netherlands) and done in 10 mL

glassy 3-neck electrochemical cells. All glassware was soaked in 5% aq. solution of

Decon 90 for 48 h followed by extensive washing with water and drying overnight in

an oven at 50 °C. All electrochemical experiments presented in this thesis were

carried out with the use of Faraday’s cage. A homemade Ag/AgCl electrode and

homemade standard calomel electrode (SCE) were used as the reference electrodes

prepared according to literature procedure.180 Platinum gauze (1 cm2) was used as the

counter electrode for the single electrode electrochemical experiments. The working

electrode was 3 mm diameter (0.071 cm2) glassy carbon rod (HTW Hochtemperatur-

Werkstoffe GmbH, Germany) sealed in glass (by Southampton University

glassblowers) and wired up with a copper wire using melted indium (Sigma-

Aldrich). Directly before electrochemical modification, blank GC electrodes were

polished with dry silicon-carbide polishing paper (grade 1200, 3M), rinsed with

acetone and sonicated in MeCN (HPLC grade) for 15 min.

4.2.2 Reagents

Linkers mono-Boc-1,2-ethylenediamine (EDA), mono-Boc-1,4-butanediamine,

mono-Boc-1,6-hexanediamine and mono-Boc-(ethylenedioxy)diethylamine (EDDA)

were purchased from Sigma-Aldrich and used without further purification. Linkers -

(N-Boc-aminomethyl)benzene diazonium tetrabluoroborate salt147,148 and tert-butyl

N-(aminomethyl)phenylmethyl)carbamate176 were prepared according to the

literature procedures. N,N-Dimethylformamide (DMF, analytical grade) was

obtained from Fisher Scienitfic and purified by distillation under reduced pressure.

Coupling reagent O-benzotriazol-1-yl)-N,N,N’,N’-tetramethylammonium

hexafluorophosphate (HBTU) was obtained from Novabiochem.

Disopropylethylamine (DIEA) (reagent grade) was purchased from Sigma-Aldrich.


149

Acetonitrile (HPLC grade) used for electrochemical experiments was obtained from

Rathburn. Aqueous solutions for electrochemical experiments were prepared using

water purified by Whatmann RO80 system coupled to a Whatmann „Still Plus”.

Homemade aqueous solution of 0.1 M phosphate buffer pH 7 was prepared using 0.1

M phosphoric acid (Fisher Scientific) and saturated solution of sodium hydroxide.

All other solvents and reagents were purchased from Sigma-Aldrich and used as

received without further purification.

4.2.3 Electrochemical modifications of GC electrodes.

3.2.1.1 General procedure for attachment of diamine linkers.181

Electrochemical attachment of mono-Boc diamine linkers (EDA, EDDA, BDA and

HDA) was performed in a solution of 20 mM mono-Boc-protected diamine and 0.1

M TBATFB in acetonitrile by electrochemical cycling (5–10 cycles) the electrode

potential in a potential range from 0 to 2.2 V vs. Ag/AgCl at a scan rate of 50 mV

s−1.

4.2.4 Attachment of (N-Boc-aminomethyl)benzene diazonium

tetrabluoroborate salt.181

Covalent attachment of diazonium salt to the GC surface was performed by

electrochemical reduction from a solution of 10 mM 4-(N-Boc-

aminomethyl)benzene diazonium tetrafluoroborate salt and 0.1 mM TBATFB in

acetonitrile. Modification of GC electrodes was carried out by cycling the electrode

potential from 0.6 to −1 V vs. Ag/AgCl for three cycles at scan rate of 50 mV s−1.


150

4.3 Solid-phase modification of the GC electrodes.

4.3.1 General procedure for Boc removal of modified GC electrodes.181

A Boc-protected modified GC electrode was suspended in a solution of HCl in

dioxane (4.0 M) at room temperature for 1 hour. The electrode was then washed in

DMF (0.5 mL), followed by subsequent washing with deionised water (0.5 mL) and

absolute EtOH (0.5 mL) before further solid-phase modification.

4.3.2 General procedure for the coupling reaction of 2,2’-bipyridine-5-

carboxylic acid at the GC surface.

2,2’-Bipyridine-5-carboxylic acid, HBTU, and DIEA were dissolved in DMF. The

mixture was gently heated for 2 minutes with a heat gun to obtain a homogenous

solution. An unprotected modified GC electrode was then suspended in this solution

which was allowed to cool to room temperature and stirred for 16 h. The electrode

was then washed with DMF, followed by absolute EtOH and allowed to dry in air for

5 minutes.

4.3.3 General procedure for coupling of the 2-(1-(carboxymethyl)-1H-

imidazol-3-ium-2-yl)pyridin-1-ium trifluoroacetate salt at the GC

electrodes.

2-(1-(Carboxymethyl)-1H-imidazol-3-ium-2-yl)pyridin-1-ium trifluoroacetate salt,

HBTU (0.46 g, 1.2 mmol) and DIEA (1.74 mL, 10 mmol) were dissolved in DMF (1

mL). The mixture was gently warmed with a heat gun to obtain a homogenous

solution. An unprotected modified GC electrode was then suspended in this solution

which was allowed to cool to room temperature and stirred for 16 h. The electrode

was subsequently washed with 1 mL of DMF followed by 1 mL of absolute EtOH

and allowed to dry in air for 5 minutes.


151

4.3.4 General procedure for grafting of ruthenium (II) complexes at the

modified GC electrodes.

The modified GC electrode with linker and binedate ligand was dipped in a solution

of the ruthenium (II) complex (27 mg, 0.05 mmol) in DMF (5 mL). The reaction

mixture was heated at 100 C° for 16 h under nitrogen. The electrode was then dipped

in 5 mL of DMF for 10 minutes followed by washing with 5 mL of EtOH and dry in

air for 5 minutes before electrochemical characterisation.

4.3.5 General procedure for grafting (1,10-phenanthroline-5,6-dione)zinc (II)

chloride complex at the GC surface.

(1,10-Phenanthroline-5,6-dione)zinc (II) chloride complex (17 mg, 0.05 mmol) was

placed in a dry multi-neck vessel and DMF (5 mL) was added under nitrogen. The

GC electrodes were dipped in the solution and heated at 50 °C overnight under

nitrogen. The electrode was dipped in 5 mL of DMF for 10 minutes followed by

washing with 5 mL of EtOH and dry in air for 5 minutes before electrochemical

characterisation.


152

4.4 General procedure for electrochemical characterisation of the

individual modified GC electrodes.

The modified electrode, SCE reference electrode and Pt counter electrode were

placed in 5 mL of 0.1 M phosphate buffer solution and degassed with argon for 5

minutes. The GC modified electrodes were characterised by cycling the potential

between −0.2 and 0.2 V vs. SCE at a scan rate of 50 mV s−1. After 10 cycles, stable

and reproducible peaks were obtained, which correspond to the redox process of the

mediator. The last cycle was saved and used for calculation of the surface coverage

of the mediator at GC electrode. Afterwards, the modified GC electrode was dipped

in 1 mM solution of NADH in aqueous solution of 0.1 M phosphate buffer solution

pH 7, which was degassed with argon for 5 minutes. The modified GC electrode was

then screened in a potential window from −0.2 to 0.2 V vs. SCE at a scan rate of 50

mV s−1. The first cycle was saved and used for analysis of the electrocatalytic

activity of the modified electrode for NADH oxidation.

The electrochemical data were reprocessed using Origin 7.0 and Origin 8.1 software.


153

4.5 Electrochemical screening of the library of 63 modified

electrodes.

High-throughput electrochemical experiments were carried out using a high-

throughput electrochemistry analyser HTEA Mark II Potentiostat (ILIKA). The HTP

electrochemical experiments were performed in a 250 mL Petri dish in homemade

0.1 M phosphate buffer solution at pH 7 as supporting electrolyte. The Petri dish was

soaked in a 5% aq. solution of Decon 90 for 48 h followed by drying overnight in an

oven at 50 °C. A homemade standard calomel electrode was used as reference

electrode and 3 cm3 Pt gauze as the counter electrode. The 0.1 M phosphate buffer

solution was degassed with argon for 5 minutes directly prior to each electrochemical

measurement.

The 63 modified electrodes were placed in a Teflon stand and dipped in 200 mL

aqueous solution of 0.1 M phosphate buffer pH 7. The solution was degassed with

argon for 15 minutes prior to each electrochemical screening. Initially, the library

was screened in a potential window from −0.3 to 0.2 V vs. SCE at a scan rate of 50

mV s−1 for 4 cycles in order to remove the residues of metal complexes physically

adsorbed at the carbon surface. Stable and reversible redox peaks were observed after

three cycles. The same procedure was repeated for screening of the library at the

following scan rates: 10, 20, 100, 200, 500 and 100 mV s−1. The solution was

degassed for 5 minutes between each measurement. The last recorded cycle for all

modified electrodes was used for evaluation of surface coverage of the mediator.

Afterwards, all electrodes in the library were dipped in a freshly prepared solution of

1 mM NADH in aqueous 0.1 M phosphate buffer pH 7 and degassed for 15 minutes.

The library was screened from −0.3 to 0.2 V vs. SCE at a scan rate of 50 mV s−1. The

same procedure was repeated for the following scan rates: 10, 100 and 200 mV s−1.

Similarly, 2 and 4 mM NADH solutions were prepared and in each case the same

screening procedure was repeated for all scan rates. The electrochemical and kinetic

analysis was performed by introducing electrochemical data in to Origin 8.1 software

and using theoretical model as described in Section 3.6.

Conclusion

154

5 Conclusion

The work presented in this thesis has demonstrated a novel strategy for the

molecular design of glassy carbon (GC) electrodes modified by various metal

complexes bearing the redox active 1,10-phenanthroline-5,6-dione ligand based on

sequential electrochemical and solid-phase synthesis methods. This strategy was

successfully applied for the preparation of a library of modified GC electrodes,

followed by HTP electrochemical screening and evaluation of the library as

electrocatalysts towards NADH oxidation.

Initial studies involved covalent functionalisation of individual GC electrodes using

sequential electrochemical attachment of different linkers followed by introduction

of different ligands and metal complexes under solid-phase conditions. The

individual modified electrodes were characterised using conventional cyclic

voltammetry towards their electrochemical and electrocatalytic properties for

oxidation of NADH. In general, the surface coverage of the covalently attached

metal complex strongly depends on the size and geometry of the metal complex.

Relatively high surface coverage was observed for tetrahedral zinc complexes at the

GC surface in comparison with the more bulky octahedral tris ruthenium complexes

bearing 2,2’-bipyridine or additional phendione chelating ligands. The metal

complexes were also attached at the surface through six different linkers in order to

study the effect of length and structure of the linker on electrochemical and

electrocatalytic activity of the modified GC electrodes. In general, calculated values

of surface coverage decreased with the length of the linker for aliphatic linkers.

Relatively low values of coverage were obtained for metal complexes attached

through the rigid benzylamine linker.

Preliminary studies on the individual modified electrodes towards electrocatalytic

activity for NADH oxidation showed that different metal complexes and different

types of linkage affect the electrocatalytic activity of the phendione redox active

ligand(s). In general, a decrease in overvoltage was observed for the oxidation of

NADH for all of the modified electrodes. An enhanced NADH electrocatalytic

activity was observed for electrodes modified with octahedral ruthenium complexes

bearing additional bidentate ligands with conjugated aromatic systems. In the case of

Conclusion

155

the aliphatic linkers, a decrease in the NADH catalytic activity of the modified

electrodes was observed with increase in length of the linkers. Relatively low values

of catalytic currents were recorded for the rigid benzylamine linker, which might

limit the phendione ligand access to the bulky NADH molecule in solution.

Based on the initial results obtained for individual electrodes, a library of 63

GC electrodes modified by different linkers, ligands and metal complexes was

designed and prepared in a combinatorial and parallel way. HTP electrochemical

screening of the library using a multichannel potentiostat allowed instant comparison

of different modification in the library during single set of measurements. An initial

result of HTP screening showed that the highest NADH catalytic currents were

recorded for modifications with the zinc complex attached at the surface through

ethylenediamine linker and this mediator system would be the best choice to apply

for activity-based biosensors. The experimental data extracted from HTP screening

of the library were also applied for evaluation of heterogeneous rate constants of the

NADH catalytic reaction according to the kinetic model reported for electrocatalysis

at different modified electrodes. Based on the catalytic rate constants, kcat, the most

effective mediator system was modification with the tris ruthenium complex attached

at the surface through the ethylenediethoxydiamine linker.

In conclusion, the results obtained for the modified GC electrodes confirmed

that the proposed strategy of sequential functionalisation of the carbon surface allows

control over the electrochemical and electrocatalytic properties of the modified

surfaces. To the best of our knowledge, the presented work is the first report

describing detailed high-throughput electrochemical and kinetic analysis towards

development of novel modified electrodes for biosensor application.

The successful preparation of array of covalently modified GC electrodes using

combinatorial methodology and the HTP screening can be applied for different redox

systems towards electrocatalytic oxidation of NADH and other substrates.

The molecular design of the carbon surface proposed in this thesis allowed for

control over the surface coverage and NADH electrocatalytic activity by introduction

of different linkers and metal complexes containing redox active phendione ligands.

In order to investigate in more detail the effect of environment on the catalytic

activity of the redox centre, a number of redox centres at the GC surface would be

decreased in a controlled manner. This can be achieved by formation of partial

Conclusion

156

coverage of different redox centres using blocking groups with different

functionalities and applying the well-known solid-phase synthesis and

electrochemical methodologies.

References

157

6 References

(1) Bunin, B. A. The Combinatorial Chemistry; Academic Press: San Francisco,

1998.

(2) Jandeleit, B.; Schaefer, D. J.; Powers, T. S.; Turner, H. W.; Weinberg, W. H.

Angew. Chem. Int. Ed. 1999, 38, 2494.

(3) Nicolaou, K. C. H., R.; Hartwig, W. Handbook of Combinatorial Chemistry;

Wiley-VCH: Weinheim, 2002; Vol. 2.

(4) Terrett, N. K. Combinatorial Chemistry; Oxford Chemistry Press Inc.: New

York, 1998.

(5) Hanak, J. J. J. Mater. Sci. 1970, 5, 964.

(6) Furka, Á.; SebestyÉN, F.; Asgedom, M.; DibÓ, G. Int. J. Pept. Protein Res.

1991, 37, 487.

(7) Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149.

(8) Ronald, F. Tetrahedron 1992, 48, 9217.

(9) Jung, G. Combinatorial Chemistry, Synthesis, Analysis, Screening; Wiley-

VCH: Weinheim, 1999.

(10) Pinilla, C.; Appel, J. R.; Blondelle, S. E.; Dooley, C. T.; Eichler, J.; Ostresh, J.

M.; Houghten, R. A. Drug Dev. Res. 1994, 33, 133.

(11) Richard A, H. Gene 1993, 137, 7.

(12) Dooley, C. T.; Chung, N. N.; Schiller, P. W.; Houghten, R. A. Proc. Natl.

Acad. Sci. 1993, 90, 10811.

(13) Houghten, R. A.; Pinilla, C.; Blondelle, S. E.; Appel, J. R.; Dooley, C. T.;

Cuervo, J. H. Nature 1991, 354, 84.

References

158

(14) Konings, D. A. M.; Wyatt, J. R.; Ecker, D. J.; Freier, S. M. J. Med. Chem.

1997, 40, 4386.

(15) Campbell, D. A.; Bermak, J. C.; Burkoth, T. S.; Patel, D. V. J. Am. Chem. Soc.

1995, 117, 5381.

(16) Vetter, D.; Tate, E. M.; Gallop, M. A. Bioconjugate Chem. 1995, 6, 319.

(17) Yudin, A. K.; Siu, T. Curr. Opinion Chem. Biol. 2001, 5, 269.

(18) Akporiaye, D. E.; Dahl, I. M.; Karlsson, A.; Wendelbo, R. Angew. Chem. Int.

Ed. 1998, 37, 609.

(19) Baker, B. E.; Kline, N. J.; Treado, P. J.; Natan, M. J. J. Am. Chem. Soc. 1996,

118, 8721.

(20) Danielson, E.; Devenney, M.; Giaquinta, D. M.; Golden, J. H.; Haushalter, R.

C.; McFarland, E. W.; Poojary, D. M.; Reaves, C. M.; Weinberg, W. H.; Wu, X. D.

Science 1998, 279, 837.

(21) Danielson, E.; Golden, J. H.; McFarland, E. W.; Reaves, C. M.; Weinberg, W.

H.; Wu, X. D. Nature 1997, 389, 944.

(22) Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Chem. Rev. 1991, 91,

1721.

(23) Burgess, K.; Lim, H.-J.; Porte, A. M.; Sulikowski, G. A. Angew. Chem. Int. Ed.

1996, 35, 220.

(24) Cole, B. M.; Shimizu, K. D.; Krueger, C. A.; Harrity, J. P. A.; Snapper, M. L.;

Hoveyda, A. H. Angew. Chem. Int. Ed. 1996, 35, 1668.

(25) Lam, K. S.; Salmon, S. E.; Hersh, E. M.; Hruby, V. J.; Kazmierski, W. M.;

Knapp, R. J. Nature 1991, 354, 82.

(26) Shimizu, K. D.; Snapper, M. L.; Hoveyda, A. H. Chem. Eur. J. 1998, 4, 1885.

References

159

(27) Francis, M. B.; Jamison, T. F.; Jacobsen, E. N. Curr. Opinion Chem. Biol.

1998, 2, 422.

(28) Bein, T. Angew. Chem. Int. Ed. 1999, 38, 323.

(29) Ohlmeyer, M. H.; Swanson, R. N.; Dillard, L. W.; Reader, J. C.; Asouline, G.;

Kobayashi, R.; Wigler, M.; Still, W. C. Proc. Natl. Acad. Sci. 1993, 90, 10922.

(30) Reddington, E. Science 1998, 280, 1735.

(31) Chen, G.; Delafuente, D. A.; Sarangapani, S.; Mallouk, T. E. Catalysis Today

2001, 67, 341.

(32) Morris, N. D.; Mallouk, T. E. J. Am. Chem. Soc. 2002, 124, 11114.

(33) Sun, Y.; Buck, H.; Mallouk, T. E. Anal. Chem. 2001, 73, 1599.

(34) Sullivan, M. G.; Utomo, H.; Fagan, P. J.; Ward, M. D. Anal. Chem. 1999, 71,

4369.

(35) Hintsche, R.; Albers, J.; Bernt, H.; Eder, A. Electroanalysis 2000, 12, 660.

(36) Spong, A. D.; Vitins, G.; Guerin, S.; Hayden, B. E.; Russell, A. E.; Owen, J. R.

J. Power Sources 2003, 119-121, 778.

(37) Guerin, S.; Hayden, B. E.; Lee, C. E.; Mormiche, C.; Owen, J. R.; Russell, A.

E.; Theobald, B.; Thompsett, D. J. Comb. Chem. 2003, 6, 149.

(38) Cooper, J. S.; McGinn, P. J. J. Power Sources 2006, 163, 330.

(39) Murray, R. M. Chemically Modified Electrodes; Marcel Dekker, Inc.: New

York, 1984; Vol. 13.

(40) Tagliazucchi, M. C., E. Chemically Modified Electrodes; Wiley-VCH, 2009;

Vol. 11.

(41) Scheller, F. W.; Wollenberger, U.; Lei, C.; Jin, W.; Ge, B.; Lehmann, C.;

Lisdat, F.; Fridman, V. Rev. Mol. Biotech. 2002, 82, 411.

References

160

(42) Collman, J. P.; Denisevich, P.; Konai, Y.; Marrocco, M.; Koval, C.; Anson, F.

C. J. Am. Chem. Soc. 1980, 102, 6027.

(43) Simon, E.; Bartlett, P. N. In Biomolecular Films: Design, Function, and

Applications; Rusling, J. F., Ed.; Marcel Dekker: New York, 2003, p 499.

(44) Jiang, R.; Chu, D. J. Electroanal. Chem. 2002, 527, 137.

(45) Downard, A. J.; Roddick, A. D.; Bond, A. M. Anal. Chim. Acta 1995, 317,

303.

(46) Kishioka, S.-y.; Ohki, S.; Ohsaka, T.; Tokuda, K. J. Electroanal. Chem. 1998,

452, 179.

(47) Zoski, C. E. Handbook of Electrochemistry; Elsevier: Amsterdam, 2007.

(48) Pilloud, D. L.; Chen, X.; Dutton, P. L.; Moser, C. C. J. Phys. Chem. B 2000,

104, 2868.

(49) Hong, H.-G.; Park, W. Langmuir 2001, 17, 2485.

(50) Laforgue, A.; Addou, T.; Bélanger, D. Langmuir 2005, 21, 6855.

(51) Paulik, M. G.; Brooksby, P. A.; Abell, A. D.; Downard, A. J. J. Phys. Chem. C

2007, 111, 7808.

(52) McCreery, R. L. Chem. Rev. 2008, 108, 2646.

(53) Heller, A.; Feldman, B. Chem. Rev. 2008, 108, 2482.

(54) Murray, R. W. Acc. Chem. Res. 1980, 13, 135.

(55) Chrétien, J.-M.; Ghanem, M. A.; Bartlett, P. N.; Kilburn, J. D. Chem. Eur. J.

2008, 14, 2548.

(56) Ghanem, M. A.; Chrétien, J.-M.; Kilburn, J. D.; Bartlett, P. N. Bioelectrochem.

2009, 76, 115.

(57) Ghanem, M. A.; Chrétien, J.-M.; Pinczewska, A.; Kilburn, J. D.; Bartlett, P. N.

J. Mater. Chem. 2008, 18, 4917.

References

161

(58) Evans, J. F.; Kuwana, T. Anal. Chem. 1977, 49, 1632.

(59) Strelko, V.; Malik, D. J. J. Colloid Interface Sci. 2002, 250, 213.

(60) Yacynych, A. M.; Kuwana, T. Anal. Chem. 1978, 50, 640.

(61) Barbier, B.; Pinson, J.; Desarmot, G.; Sanchez, M. J. Electrochem. Soc. 1990,

137, 1757.

(62) Deinhammer, R. S.; Ho, M.; Anderegg, J. W.; Porter, M. D. Langmuir 1994,

10, 1306.

(63) Andrieux, C. P.; Gonzalez, F.; Savéant, J.-M. J. Am. Chem. Soc. 1997, 119,

4292.

(64) Delamar, M.; Hitmi, R.; Pinson, J.; Saveant, J. M. J. Am. Chem. Soc. 1992,

114, 5883.

(65) Downard, A. J. Electroanalysis 2000, 12, 1085.

(66) Adenier, A.; Chehimi, M. M.; Gallardo, I.; Pinson, J.; Vilà, N. Langmuir 2004,

20, 8243.

(67) Maeda, H. Y., Y.; Hosoe, M.; Li, T.-X.; Yamaguchi, E.; Kasamatsu, M.;

Ohmori, H.; Chem. Pharm Bull. 1994, 42, 1870.

(68) Gooding, J. J. Electroanalysis 2008, 20, 573.

(69) Saby, C.; Ortiz, B.; Champagne, G. Y.; Bélanger, D. Langmuir 1997, 13, 6805.

(70) Anariba, F.; Steach, J. K.; McCreery, R. L. J. Phys. Chem. B 2005, 109, 11163.

(71) Liu, G.; Liu, J.; Böcking, T.; Eggers, P. K.; Gooding, J. J. Chemical Physics

2005, 319, 136.

(72) Liu, G.; Liu, J.; Davis, T. P.; Gooding, J. J. Biosensors and Bioelectronics

2011, 26, 3660.

(73) Pinson, J.; Podvorica, F. Chem. Soc. Rev. 2005, 34, 429.

(74) Kariuki, J. K.; McDermott, M. T. Langmuir 2001, 17, 5947.

References

162

(75) Anariba, F.; DuVall, S. H.; McCreery, R. L. Anal. Chem. 2003, 75, 3837.

(76) Bugg, T. Introduction to Enzyme and Coenzyme Chemistry; Blackwell

Publishing Ltd., 2004.

(77) Groton, L.; Bartlett, P. N. Bioelectrochemistry; Fundamentals, Experimental

Technqiues and Applications; Wiley, 2008.

(78) Tanner, J. J. T., S.; Barbour, C. L. J.; Barnes, C. L.; Krause, K. L. Protein

Science 1999, 8, 1725.

(79) Loewus, F. A.; Ofner, P.; Fisher, H. F.; Westheimer, F. H.; Vennesland, B. J

Biol Chem 1953, 202, 699.

(80) Loewus, F. A.; Westheimer, F. H.; Vennesland, B. J. Am. Chem. Soc. 1953, 75,

5018.

(81) Miles, D. W. U., D. W. J. Biol. Chem. 1968, 243, 4181.

(82) Smith, P. E. J. Am. Chem. Soc. 1999, 121, 8637.

(83) Bell, C. E. Y., T. O.; Eisenberg, D. Protein Science 1997, 2084.

(84) Rossman, M. G. L., Anders; Branden, Carl-Ivar; Banaszak, Leonard J.

Evolutionaly and Structurally Relationship among Dehydrogenases; Academic

Press: Los Angeles; Vol. 11.

(85) Branden, C. T., John Introduction to protein structure; Second ed.; Garland

Publishing, Inc.: New York, 1999.

(86) Arthur M, L. Curr. Opin. Struct. Biol. 1995, 5, 775.

(87) Brändén, C.-I. Q. Rev. Biophys. 1980, 13, 317.

(88) Lobo, M. J.; Miranda, A. J.; Tuñón, P. Electroanalysis 1997, 9, 191.

(89) Devaux-Basseguy, R.; Bergel, A.; Comtat, M. Enzyme Microb. Tech. 1997, 20,

248.

(90) Rodkey, F. L. J Biol Chem 1955, 213, 777.

References

163

(91) Clark, W. M. Oxidation-Reduction Potentials of Organic Systems; Robert E.

Krieger Publishing: Huntington, 1972.

(92) Burnett, J. N.; Underwood, A. L. Biochemistry 1965, 4, 2060.

(93) Samec, Z. E., Philip J. J. Electroanal. Chem. 1983, 144, 217.

(94) Blaedel, W. J.; Jenkins, R. A. Anal. Chem. 1975, 47, 1337.

(95) Braun, R. D.; Santhanam, K. S. V.; Elving, P. J. J. Am. Chem. Soc. 1975, 97,

2591.

(96) Blankespoor, R. L. M., Larry L. J. Electroanal. Chem. 1984, 171, 231.

(97) Gorton, L. J. Chem. Soc., Faraday Trans. 1986, 82, 1245.

(98) Moiroux, J.; Elving, P. J. J. Am. Chem. Soc. 1980, 102, 6533.

(99) Tse, D. C.-S.; Kuwana, T. Anal. Chem. 1978, 50, 1315.

(100) Jaegfeldt, H. Bioelectrochem. Bioelectron. 1981, 8, 355.

(101) Elving, P. J. B., William T.; Morioux, Jacques; Samec, Zdenek

Bioelectrochem. Bioelectron. 1982, 9, 365.

(102) Bartlett, P. N.; Wallace, E. N. K. J. Electroanal. Chem. 2000, 486, 23.

(103) Jaegfeldt, H.; Kuwana, T.; Johansson, G. J. Am. Chem. Soc. 1983, 105, 1805.

(104) Bartlett, P. N. T., P.; Whitaker, R. G. Prog. React. Kinet. 1991, 16, 55.

(105) Gorton, L.; Dominguez, E. In Encyclopedia of Electrochemistry; Bard, A. J.,

Stratmann, M., Eds.; Wiley-VCH Verlag GmbH & Co.: Weinheim, 2002; Vol. 9, p

69.

(106) Bartlett, P. N.; Birkin, P. R.; Wallace, E. N. K. J. Chem. Soc., Faraday Trans.

1997, 93, 1951.

(107) Kitani, A.; So, Y. H.; Miller, L. L. J. Am. Chem. Soc. 1981, 103, 7636.

(108) Gorton, L.; Torstensson, A.; Jaegfeldt, H.; Johannson, G. J. Electroanal. Chem.

1984, 161, 103.

References

164

(109) Pariente, F.; Lorenzo, E.; Abruna, H. D. Anal. Chem. 1994, 66, 4337.

(110) Ni, F.; Feng, H.; Gorton, L.; Cotton, T. M. Langmuir 1990, 6, 66.

(111) Cosnier, S.; Le Lous, K. J. Electroanal. Chem. 1996, 406, 243.

(112) Chen, H.-Y.; Zhou, D.-M.; Xu, J.-J.; Fang, H.-Q. J. Electroanal. Chem. 1997,

422, 21.

(113) Ohtani, M.; Kuwabata, S.; Yoneyama, H. J. Electroanal. Chem. 1997, 422, 45.

(114) Bartlett, P. N.; Simon, E. Phys. Chem. Chem. Phys. 2000, 2, 2599.

(115) Schuhmann, W.; Lammert, R.; Hämmerle, M.; Schmidt, H.-L. Biosensors and

Bioelectronics 1991, 6, 689.

(116) Wu, Q.; Maskus, M.; Pariente, F.; Tobalina, F.; Fernández, V. M.; Lorenzo, E.;

Abruña, H. D. Anal. Chem. 1996, 68, 3688.

(117) Ju, H.; Leech, D. Anal. Chim. Acta 1997, 345, 51.

(118) Storrier, G. D.; Takada, K.; Abruña, H. D. Inorg. Chem. 1999, 38, 559.

(119) Gaspar, S.; Habermüller, K.; Csöregi, E.; Schuhmann, W. Sensors and

Actuators B: Chemical 2001, 72, 63.

(120) Albery, W. J.; Bartlett, P. N. J. Chem. Soc., Chem. Comm. 1984, 234.

(121) Cai, C. X.; Xue, K. H. Microchem. J. 1998, 58, 197.

(122) Mano, N.; Kuhn, A. J. Electroanal. Chem. 1999, 477, 79.

(123) Casero, E.; Darder, M.; Takada, K.; Abruña, H. D.; Pariente, F.; Lorenzo, E.

Langmuir 1998, 15, 127.

(124) Eckert, T. S.; Bruice, T. C. J. Am. Chem. Soc. 1983, 105, 4431.

(125) Evans, D. H. G., D. A. J. Electroanal. Chem. 1982, 134, 301.

(126) Hilt, G.; Steckhan, E. J. Chem. Soc., Chem. Comm. 1993, 1706.

(127) Goss, C. A.; Abruna, H. D. Inorg. Chem. 1985, 24, 4263.

(128) Inglett, G. E.; Smith, G. F. J. Am. Chem. Soc. 1950, 72, 842.

References

165

(129) Eckert, T. S.; Bruice, T. C.; Gainor, J. A.; Weinreb, S. M. Proc. Natl. Acad.

Sci. 1982, 79, 2533.

(130) Rivera, N.; Colón, Y.; Guadalupe, A. R. Bioelectrochem. Bioenerget. 1994, 34,

169.

(131) Hedenmo, M.; Narvaez, A.; Dominguez, E.; Katakis, I. Analyst 1996, 121,

1891.

(132) Santiago, M. E. B.; Vélez, M. M.; Borrero, S.; Díaz, A.; Casillas, C. A.;

Hofmann, C.; Guadalupe, A. R.; Colón, J. L. Electroanalysis 2006, 18, 559.

(133) Tobalina, F.; Pariente, F.; Hernández, L.; Abruña, H. D.; Lorenzo, E. Anal.

Chim. Acta 1999, 395, 17.

(134) Catalin Popescu, I.; DomInguez, E.; Narváez, A.; Pavlov, V.; Katakis, I. J.

Electroanal. Chem. 1999, 464, 208.

(135) Dixon, M.; Webb, E. C. Enzymes; Longman: London, 1979.


2009, 15, 11928.

(137) Evans, I. P.; Spencer, A.; Wilkinson, G. J. Chem. Soc. Dalton Trans. 1973,

204.

(138) Yamada, M.; Tanaka, Y.; Yoshimoto, Y.; Kuroda, S.; Shimao, I. Bull. Chem.

Soc. Jpn. 1992, 65, 1006.

(139) Hudali, H. A.; Kingston, J. V.; Tayim, H. A. Inorg. Chem. 1979, 18, 1391.

(140) Alessio, E. Chem. Rev. 2004, 104, 4203.

(141) Lever, A. B. P. Inorg. Chem. 1990, 29, 1271.

(142) Zakeeruddin, S. M.; Nazeeruddin, M. K.; Humphry-Baker, R.; Gratzel, M.;

Shklover, V. Inorg. Chem. 1998, 37, 5251.

(143) Chung, C. W. Y.; Toy, P. H. J. Comb. Chem. 2007

References

166

9, 115.

(144) Jiang, B.; Hao, W.-J.; Wang, X.; Shi, F.; Tu, S.-J. J. Comb. Chem. 2009, 11,

846.

(145) Boland, S.; Barrière, F.; Leech, D. Langmuir 2008, 24, 6351.

(146) Blaedel, W. J.; Haas, R. G. Anal. Chem. 1970, 42, 918.

(147) Lee, J.; Lee, J.; Kang, M.; Shin, M.; Kim, J.-M.; Kang, S.-U.; Lim, J.-O.; Choi,

H.-K.; Suh, Y.-G.; Park, H.-G.; Oh, U.; Kim, H.-D.; Park, Y.-H.; Ha, H.-J.; Kim, Y.-

H.; Toth, A.; Wang, Y.; Tran, R.; Pearce, L. V.; Lundberg, D. J.; Blumberg, P. M. J.

Med. Chem. 2003, 46, 3116.

(148) McNab, H.; Monahan, L. C. J. Chem. Soc., Perkin Trans. 1 1989, 419.

(149) Hoekstra, K. J.; Bein, T. Chem Mater 1996, 8, 1865.

(150) Boghaei, D. M.; Behzadian-Asl, F. J. Coord. Chem. 2007, 60, 347

(151) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. Handbook X-Ray

Photoelectron Spectroscopy Perkin-Elmer Corporation: Minnesota, US, 1992.

(152) Huang, L.; Wang, K.-Z.; Huang, C.-H.; Li, F.-Y.; Huang, Y.-Y. J. Mater.

Chem. 2001, 11, 790.

(153) Leirer, M.; Knör, G.; Vogler, A. Inorg. Chim. Acta 1999, 288, 150.

(154) Yi, H.; Crayston, J. A.; Irvine, J. T. S. Dalton Trans. 2003, 685.

(155) Ackermann, Y.; Guschin, D. A.; Eckhard, K.; Shleev, S.; Schuhmann, W.

Electrochem. Comm. 2010, 12, 640.

(156) Wagner, H. J.; Loutfy, R. O.; Hsiao, C.-K. J. Mater. Sci. 1982, 17, 2781.

(157) Yue, S.-M.; Xu, H.-B.; Ma, J.-F.; Su, Z.-M.; Kan, Y.-H.; Zhang, H.-J.

Polyhedron 2006, 25, 635.

(158) Huan, Z.; Persson, B.; Gorton, L.; Sahni, S.; Skotheim, T.; Bartlett, P.

Electroanalysis 1996, 8, 575.

References

167

(159) Gorton, L. J. Chem. Soc., Perkin Trans. 1 1986, 82, 1245.

(160) Gorton, L.; Johansson, G.; Torstensson, A. J. Electroanal. Chem. 1985, 196,

81.

(161) Albery, W. J.; Bartlett, P. N.; Lithgow, A. M.; Riefkohl, J. L.; Rodriguez, L.

A.; Romero, L.; Souto, F. A. J. Org. Chem. 1985, 50, 596.

(162) Albery, W. J.; Bartlett, P. N.; Wilde, C. P.; Darwent, J. R. J. Am. Chem. Soc.

1985, 107, 1854.

(163) Albery, W. J.; Knowles, J. R. Biochem. 1976, 15, 5627.



(166) Albery, W. J.; Bartlett, P. N. J. Electroanal. Chem. 1985, 194, 211.

(167) Albery, J. W.; Bartlett, P. N.; Craston, D. H. J. Electroanal. Chem. 1985, 194,

223.

(168) Lyons, M. E. G.; Lyons, C. H.; Michas, A.; Bartlett, P. N. J. Electroanal.

Chem. 1993, 351, 245.

(169) Bartlett, P. N.; Wallace, E. N. K. Phys. Chem. Chem. Phys. 2001, 3, 1491.

(170) Lyons, M. E. G.; Fitzgerald, C. A.; Smyth, M. R. Analyst 1994, 119, 855.

(171) Albery, J. Electrode kinetics; Clarendon Press: Oxford, 1975.

(172) Zanello, P. Inorganic Electrochemistry; Theory, Practice and Application;

Royal Society of Chemistry, 2003.

(173) Wallace, E. N. K. Ph.D. Thesis, University of Southampton, 1997.

(174) Bartlett, P. N.; Birkin, P. R.; Wallace, E. N. K. J. Chem. Soc., Faraday Trans.

1997, 93, 1951.

(175) Yokoyama, K.; Ueda, Y.; Nakamura, N.; Ohno, H. Chem. Lett. 2005, 34, 1282.

References

168

(176) Goodyer, C. L. M.; Chinje, E. C.; Jaffar, M.; Stratford, I. J.; Threadgill, M. D.

Biorg. Med. Chem. 2003, 11, 4189.

(177) M. Yamada, Y. T., Y. Yoshimoto, S. Kuroda, J. Shimaro Bull. Chem. Soc. Jpn.

1992, 65, 1006.

(178) Evans, I. P.; Spencer, A.; Wilkinson, G. J. Chem. Soc., Dalton Trans. 1973,

204.

(179) Akasheh, T. S.; Marji, D.; Al-Ahmed, Z. M. Inorg. Chim. Acta 1988, 141, 125.

(180) Bartlett, P. N. Biosensors: A Practical Approach; 1st ed.; Oxford University

Press: Oxford, 1990.


2008, 14, 2548.

Documents

University of Southampton Research Repository …...University of Southampton Faculty of Natural and Environmental Sciences School of Chemistry Doctor of Philosophy High-throughput